Servo two-degree-of-freedom control method based on RST control structure
Through the servo two-degree-of-freedom control method based on the RST control structure, the tracking and interference suppression transfer function is independently designed, which solves the problem of taking into account both performance of traditional controllers in higher-order systems, and achieves more refined and accurate control.
Patent Information
- Application Number
- CN202411228023.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-03
- Publication Date
- 2024-12-06
AI Technical Summary
The traditional unit feedback closed-loop control loop cannot independently adjust the tracking and interference suppression transfer functions during design, which makes it difficult to take into account the time domain performance and interference suppression, and the controller parameters are insufficient to calm the high-order controlled system, and there is error accumulation and Challenges to predict future signals.
Using the servo two-degree of freedom control method based on the RST control structure, the tracking and interference suppression transfer function is independently designed through data acquisition, interference suppression, tracking dynamic setting and control algorithm implementation module, and zero steady-state error compensation and wave limit filter are added. Design the controller directly in the discrete domain to avoid accumulation and predicted signal processing.
It realizes independent design tracking and interference suppression dynamics, fully considers the dynamic characteristics of higher-order systems, provides more refined control, simplifies the controller implementation, and improves control accuracy and stability.
Smart Images

Figure CN119087793A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of servo control, and in particular relates to a servo two-degree-of-freedom control method based on an RST control structure. Background Art
[0002] Disadvantages of traditional unit feedback closed-loop control loop:
[0003] The tracking transfer function in the traditional unit feedback closed-loop control loop determines the system's time domain performance, such as bandwidth, overshoot, and settling time, while the transfer function from disturbance to output determines the disturbance suppression performance. When designing the tracking transfer function, the controller parameters are uniquely determined, which results in the transfer functions from disturbance to output and disturbance to control quantity being unable to be changed. As a result, the system can only compromise between time domain performance and disturbance suppression, sacrificing each other's performance. If the disturbance is included in the control loop bandwidth, a filter must be added to the tracking transfer function, which will lead to a significant decrease in tracking performance (time domain performance).
[0004] In servo control, traditional controllers usually have only two parameters: proportional gain and integral gain. The controller contains a limited number of zeros and poles, making it difficult to stabilize high-order controlled systems and system dynamics, and even more difficult to meet higher requirements. The industry mainly uses TI's controller design method, which approximates the bandwidth of the closed-loop transfer function by the crossover frequency of the open-loop transfer function of the speed loop, thereby designing a speed loop controller. However, in the application of non-direct-drive loads such as harmonic reducers, the harmonics themselves have damping and spring stiffness, and their true open-loop transfer function is quite different from the result calculated by the traditional method. The closed-loop transfer function of the servo-side speed loop has a higher order, and the traditional controller has only two parameters, which cannot fully consider the complexity of high-order systems.
[0005] Traditional technologies have too few controller parameters to fully consider the complexity of high-order systems.
[0006] Traditional controllers involve integrators, which need to store and accumulate all past errors, which may lead to excessive error accumulation. The introduction of differentiators requires the prediction of future signals, which increases the complexity and challenges of implementation. Traditional unit feedback closed-loop control loops have significant disadvantages in terms of difficulty in balancing time domain performance and interference suppression, difficulty in stabilizing high-order controlled systems, insufficient controller parameters and optimization difficulties, and challenges in accumulating errors and predicting future signals. These problems limit the application of traditional controllers in high-precision and complex systems. Summary of the invention
[0007] The purpose of the present invention is to provide a servo two-degree-of-freedom control method based on an RST control structure, which solves the technical problem of independently designing tracking and interference suppression transfer functions in the control process of a controller.
[0008] To achieve the above object, the present invention adopts the following technical solution:
[0009] A servo two-degree-of-freedom control method based on an RST control structure comprises the following steps:
[0010] Step 1: The data acquisition module acquires input data, which includes a speed given value and a speed output; the transfer function module uses the input data to define the controlled target transfer function of the speed loop and converts it into a discrete transfer function;
[0011] Step 2: The interference suppression module sets the adjustment dynamics, designs the transfer function from interference to output, adds compensation for zero steady-state error, and removes interference by setting bandwidth or adding a wave-limiting filter;
[0012] Step 3: The tracking dynamics module sets the ideal tracking dynamics from the reference input to the output. It calculates and generates the controller polynomial by setting the dynamic characteristics and stability margin.
[0013] Step 4: The control algorithm implementation module implements the control algorithm according to the polynomial calculation results of steps 2 and 3, generates the control quantity, checks the dynamic response of the control loop through the simulation environment, and makes fine adjustments;
[0014] Step 5: The joint reducer module calculates the transfer function of the q-axis current given to the output speed while considering the joint reducer, and converts it into a discrete form. The order of the controller is adjusted to consider the influence of the resonant frequency and the anti-resonant frequency, and then the control algorithm is improved by repeating the process from step 1 to step 4.
[0015] Preferably, when executing step 1, the following steps are specifically included:
[0016] Step 1-1: Assume that r(t) is the reference input, i.e. the speed given value, y(t) is the speed output actually collected, i.e. the output, v(t) is the interference signal, and u(t) is the control quantity given to the controlled target of the speed loop;
[0017] Find the controlled target transfer function G(s) of the speed loop:
[0018]
[0019] Where G(s) is the transfer function from the speed controller output to the speed feedback, current_closed_loop represents the current loop transfer function, P represents the number of motor poles, and ψ f represents the rotor flux of the PMSM motor, J represents the total moment of inertia of the motor, and s is the Laplace operator;
[0020] Step 1-2: Convert the transfer function G(s) into a discrete transfer function And use MATLAB to find the delay d to get the controlled target G(q -1 ):
[0021]
[0022] Wherein, q represents the forward shift operator.
[0023] Preferably, when executing step 2, the following steps are specifically included:
[0024] Step 2-1: Design the transfer function S_sensitivity from disturbance v(t) to output y(t):
[0025]
[0026] Among them, A(q -1 ) is the denominator polynomial of the speed loop controlled target, B(q -1 ) is the numerator polynomial of the controlled target of the speed loop, S(q -1 ) and R(q -1 ) are two polynomials of the RST controller, q -d is the delay operator, used to represent the delay of the controlled system;
[0027] Set the denominator polynomial P(q -1 ):
[0028] A(q -1 )H S (q -1 )S′(q -1 )+q -d B(q -1 )H R (q -1 )R′(q -1 )=P(q -1 );
[0029] Among them, S(q -1 )=HS(q -1 )S′(q -1 )R(q -1 )=H R (q -1 )R(q -1 ).
[0030] Step 2-2: Add H S (q -1 )=1-q -1 To obtain zero steady-state error, add H R (q -1 )=1+q -1 Open the loop above the Nyquist frequency to filter out high-frequency interference;
[0031] Step 2-3: Calculate S′(q -1 ) and R′(q -1 ) and obtain S(q -1 ) and R(q -1 )of.
[0032] Preferably, when executing step 3, the following steps are specifically included:
[0033] Step 3-1: Design the reference input to output transfer function H cl (q -1 ):
[0034]
[0035] Step 3-2: By T(q -1 ) to select, thereby eliminating the tracking dynamics P(q -1 );
[0036] Step 3-3: Get The polynomial coefficients are calculated and set to complete the controller-related parameter calculations.
[0037] Preferably, when executing step 4, the following steps are specifically included:
[0038] Step 4-1: Collect the given speed r(t) and use the polynomial T(q -1 ) Generate a signal;
[0039] Step 4-2: Through R(q -1 ) obtain the feedback signal of the speed closed-loop output y(t);
[0040] Step 4-3: After the signal is subtracted, pass 1 / S(q -1 ) Generate the control quantity u(t);
[0041] Step 4-4: Check the dynamic response in the simulation environment and adjust the damping ratio and natural frequency of the transfer function.
[0042] Preferably, when executing step 5, the following steps are specifically included:
[0043] Step 5-1: Calculate the transfer function P1(s) from the joint driving torque to the output end velocity:
[0044]
[0045] Where s represents the Laplace operator, ω n and ω z denote the resonant frequency and anti-resonant frequency respectively, ξ n and zRespectively represent the damping ratios corresponding to the resonant frequency and the anti-resonant frequency;
[0046] Finally, the transfer function P1(s) is divided by the motor torque constant to obtain the transfer function from the motor q-axis current to the speed loop speed output;
[0047] Step 5-2: Increase the order of the controller to take into account the resonant frequency and anti-resonant frequency;
[0048] Step 5-3: Repeat the process from step 1 to step 4 to improve the control algorithm.
[0049] The servo two-degree-of-freedom control method based on the RST control structure described in the present invention solves the technical problem of independently designing the dynamics of tracking and interference suppression rather than compromising the two during the control process of the controller. The present invention can fully consider the dynamic characteristics of the high-order system, directly and accurately specify the closed-loop dynamic performance and stability margin of the entire servo speed loop, rather than using approximate means, to provide more refined control, and can easily take into account the complex load dynamics outside the motor shaft. The controller itself is designed directly in the discrete domain, and the controller does not have a continuous to discrete process, and does not need to accumulate past signals or predict future signals, which simplifies the implementation of the controller and effectively responds through independent design and high-order control, thereby achieving more precise control. BRIEF DESCRIPTION OF THE DRAWINGS
[0050] Figure 1 It is the main flow chart of the present invention;
[0051] Figure 2 This is a rendering of the present invention after adding fixed items;
[0052] Figure 3 It is a control schematic diagram of step 3 of the present invention. DETAILED DESCRIPTION
[0053] Depend on Figure 1-Figure 3 A servo two-degree-of-freedom control method based on an RST control structure is shown, comprising the following steps:
[0054] Step 1: The data acquisition module acquires input data, which includes a speed given value and a speed output; the transfer function module uses the input data to define the controlled target transfer function of the speed loop and converts it into a discrete transfer function;
[0055] In this embodiment, in the RST controller, S(q -1 )、R(q -1 ) and T(q -1 ) are the three polynomials in the controller, which are:
[0056] The polynomial processes the error signal and sends it to the controlled object as a control signal.
[0057] R(q -1 ): This polynomial processes the closed-loop system output (i.e., the output signal that needs to be controlled) and returns the difference with the reference input signal processed by the following polynomial to generate an error signal.
[0058] T(q -1 ): Used to generate an error signal by subtracting the command signal from the feedback signal after processing.
[0059]
[0060] When executing step 1, the specific steps include:
[0061] Step 1-1: Assume that r(t) is the reference input, i.e. the speed given value, y(t) is the speed output actually collected, i.e. the output, v(t) is the interference signal, and u(t) is the control quantity given to the controlled target of the speed loop;
[0062] Find the controlled target transfer function G(s) of the speed loop:
[0063]
[0064] Where G(s) is the transfer function from the speed controller output to the speed feedback, current_closed_loop represents the current loop transfer function, P represents the number of motor poles, and ψ f represents the rotor flux of the PMSM motor, J represents the total moment of inertia of the motor, and s is the Laplace operator;
[0065] Step 1-2: Convert the transfer function G(s) into a discrete transfer function And use MATLAB to find the delay d to get the controlled target G(q -1 ):
[0066]
[0067] Wherein, q represents the forward shift operator.
[0068] In this embodiment, the speed loop module obtains system parameters (such as current loop, transition frequency, number of motor poles, rotor flux, moment of inertia, etc.) and defines the transfer function G(q -1 ).
[0069] Step 2: The interference suppression module sets the adjustment dynamics, designs the transfer function from interference to output, adds compensation for zero steady-state error, and removes interference by setting bandwidth or adding a wave-limiting filter;
[0070] When executing step 2, the specific steps include:
[0071] Step 2-1: Design the transfer function S_sensitivity from disturbance v(t) to output y(t):
[0072]
[0073] Among them, A(q -1 ) is the denominator polynomial of the speed loop controlled target, B(q -1 ) is the numerator polynomial of the controlled target of the speed loop, S(q -1 ) and R(q -1 ) are two polynomials of the RST controller, q -d is the delay operator, used to represent the delay of the controlled system;
[0074] Set the denominator polynomial P(q -1 ):
[0075] A(q -1 )H S (q -1 )S′(q -1 )+q -d B(q -1 )H R (q -1 )R′(q -1 )=P(q -1 );
[0076] Among them, S(q -1 )=H S (q -1 )S′(q -1 ), R(q -1 )=H R (q -1 )R′(q -1 ).
[0077] Step 2-2: Add H S (q -1 )=1-q -1 To obtain zero steady-state error, add H R (q -1 )=1+q -1 Open the loop above the Nyquist frequency to filter out high-frequency interference; the effect is as follows Figure 2 shown.
[0078] Step 2-3: Calculate S′(q -1 ) and R′(q -1 ) and obtain S(q -1 ) and R(q -1 )of.
[0079] In this embodiment, by solving the interference suppression equation, S′(q -1 ) and R′(q -1 ), including:
[0080] S1: According to the set interference suppression target, design the transfer function P(q -1 );
[0081] S2: According to the formula
[0082] A(q -1 )H S (q -1 )S′(q -1 )+q -b B(q -1 )H R (q -1 )R′(q -1 )=P(q -1 )Calculate S′(q -1 ) and R′(q -1 ).
[0083] S3: Use the calculated S′(q -1 ) and R′(q -1 ) as the preliminary solution of the regulating polynomial and tracking polynomial, and then further adjustment and optimization are carried out in the actual design process to ensure that the system meets the design requirements. -1 ) and R(q -1 ) to achieve the target performance and verify whether the design meets the interference suppression and tracking requirements. In this embodiment, the equations can be solved by MATLAB.
[0084] Step 3: The tracking dynamics module sets the ideal tracking dynamics from the reference input to the output. It calculates and generates the controller polynomial by setting the dynamic characteristics and stability margin.
[0085] When executing step 3, the specific steps include:
[0086] Step 3-1: Design the reference input to output transfer function H cl (q -1 ):
[0087]
[0088] Step 3-2: By T(q -1 ) to select, thereby eliminating the tracking dynamics P(q -1 );
[0089] Step 3-3: Get The polynomial coefficients are calculated and set to complete the controller-related parameter calculations.
[0090] Step 4: The control algorithm implementation module implements the control algorithm according to the polynomial calculation results of steps 2 and 3, generates the control quantity, checks the dynamic response of the control loop through the simulation environment, and makes fine adjustments;
[0091] When executing step 4, the specific steps include:
[0092] Step 4-1: Collect the given speed r(t) and use the polynomial T(q -1 ) Generate a signal;
[0093] Step 4-2: Through R(q -1 ) obtain the feedback signal of the speed closed-loop output y(t);
[0094] Step 4-3: After the signal is subtracted, pass 1 / S(q -1 ) Generate the control quantity u(t);
[0095] Step 4-4: Check the dynamic response in the simulation environment and adjust the damping ratio and natural frequency of the transfer function.
[0096] Step 5: The joint reducer module calculates the transfer function of the q-axis current given to the output speed while considering the joint reducer, and converts it into a discrete form. The order of the controller is adjusted to consider the influence of the resonant frequency and the anti-resonant frequency, and then the control algorithm is improved by repeating the process from step 1 to step 4.
[0097] When executing step 5, the specific steps include:
[0098] Step 5-1: Calculate the transfer function P1(s) from the joint driving torque to the output end velocity:
[0099]
[0100] Where s represents the Laplace operator, ω n and ω z denote the resonant frequency and anti-resonant frequency respectively, ξ n and z Respectively represent the damping ratios corresponding to the resonant frequency and the anti-resonant frequency;
[0101]
[0102] Finally, the transfer function P1(s) is divided by the motor torque constant to obtain the transfer function from the motor q-axis current to the speed loop speed output;
[0103] Step 5-2: Increase the order of the controller to take into account the resonant frequency and anti-resonant frequency;
[0104] Step 5-3: Repeat the process from step 1 to step 4 to improve the control algorithm.
[0105] The servo two-degree-of-freedom control method based on the RST control structure described in the present invention solves the technical problem of independently designing the dynamics of tracking and interference suppression rather than compromising the two during the control process of the controller. The present invention can fully consider the dynamic characteristics of the high-order system, directly and accurately specify the closed-loop dynamic performance and stability margin of the entire servo speed loop, rather than using approximate means, to provide more refined control, and can easily take into account the complex load dynamics outside the motor shaft. The controller itself is designed directly in the discrete domain, and the controller does not have a continuous to discrete process, and does not need to accumulate past signals or predict future signals, which simplifies the implementation of the controller and effectively responds through independent design and high-order control, thereby achieving more precise control.
Claims
1. A servo two-degree-of-freedom control method based on an RST control structure, characterized in that: The steps include: Step 1: The data acquisition module acquires input data, which includes a speed given value and a speed output; the transfer function module uses the input data to define the controlled target transfer function of the speed loop and converts it into a discrete transfer function; Step 2: The interference suppression module sets the adjustment dynamics, designs the transfer function from interference to output, adds compensation for zero steady-state error, and removes interference by setting bandwidth or adding a wave-limiting filter; Step 3: The tracking dynamics module sets the ideal tracking dynamics from the reference input to the output. It calculates and generates the controller polynomial by setting the dynamic characteristics and stability margin. Step 4: The control algorithm implementation module implements the control algorithm according to the polynomial calculation results of steps 2 and 3, generates the control quantity, checks the dynamic response of the control loop through the simulation environment, and makes fine adjustments; Step 5: The joint reducer module calculates the transfer function of the q-axis current given to the output speed while considering the joint reducer, and converts it into a discrete form. The order of the controller is adjusted to consider the influence of the resonant frequency and the anti-resonant frequency, and then the control algorithm is improved by repeating the process from step 1 to step 4.
2. A servo two-degree-of-freedom control method based on an RST control structure as claimed in claim 1, characterized in that: When executing step 1, the specific steps include: Step 1-1: Assume that r(t) is the reference input, i.e. the speed given value, y(t) is the speed output actually collected, i.e. the output, v(t) is the interference signal, and u(t) is the control quantity given to the controlled target of the speed loop; Find the controlled target transfer function G(s) of the speed loop: Where G(s) is the transfer function from the speed controller output to the speed feedback, current_closed_loop represents the current loop transfer function, P represents the number of motor poles, and ψ f represents the rotor flux of the PMSM motor, J represents the total moment of inertia of the motor, and s is the Laplace operator; Step 1-2: Convert the transfer function G(s) into a discrete transfer function And use MATLAB to find the delay d to get the controlled target G(q -1 ): Wherein, q represents the forward shift operator.
3. A servo two-degree-of-freedom control method based on an RST control structure as claimed in claim 2, characterized in that: When executing step 2, the specific steps include: Step 2-1: Design the transfer function S_sensitivity from disturbance v(t) to output y(t): Among them, A(q -1 ) is the denominator polynomial of the speed loop controlled target, B(q -1 ) is the numerator polynomial of the controlled target of the speed loop, S(q -1 ) and R(q -1 ) are two polynomials of the RST controller, q -d is the delay operator, used to represent the delay of the controlled system; Set the denominator polynomial P(q -1 ): A(q -1 )H S (q -1 )S′(q -1 )+q -d B(q -1 )H R (q -1 )R′(q -1 )=P(q -1 ); where \(S(q -1 ) = H S (q -1 )S'(q -1 ), R(q -1 ) = H R (q -1 )R'(q -1 ). Step 2-2: Add H S (q -1 )=1-q -1 To obtain zero steady-state error, add H R (q -1 )=1+q -1 Open the loop above the Nyquist frequency to filter out high-frequency interference; Step 2-3: Calculate S′(q -1 ) and R′(q -1 ) and obtain S(q -1 ) and R(q -1 )of.
4. A servo two-degree-of-freedom control method based on an RST control structure as claimed in claim 3, characterized in that: When executing step 3, the specific steps include: Step 3-1: Design the reference input to output transfer function H cl (q -1 ): Step 3-2: By T(q -1 ) to select, thereby eliminating the tracking dynamics P(q -1 ); Step 3-3: Get The polynomial coefficients are calculated and set to complete the controller-related parameter calculation and setting.
5. A servo two-degree-of-freedom control method based on an RST control structure as claimed in claim 4, characterized in that: When executing step 4, the specific steps include: Step 4-1: Collect the given speed r(t) and use the polynomial T(q -1 ) Generate a signal; Step 4-2: Through R(q -1 ) obtain the feedback signal of the speed closed-loop output y(t); Step 4-3: After the signal is subtracted, pass 1 / S(q -1 ) Generate the control quantity u(t); Step 4-4: Check the dynamic response in the simulation environment and adjust the damping ratio and natural frequency of the transfer function.
6. A servo two-degree-of-freedom control method based on an RST control structure as claimed in claim 1, characterized in that: When executing step 5, the specific steps include: Step 5-1: Calculate the transfer function P1(s) from the joint driving torque to the output end velocity: Where s represents the Laplace operator, ω n and ω z denote the resonant frequency and anti-resonant frequency respectively, ξ n and z Respectively represent the damping ratios corresponding to the resonant frequency and the anti-resonant frequency; Finally, the transfer function P1(s) is divided by the motor torque constant to obtain the transfer function from the motor q-axis current to the speed loop speed output; Step 5-2: Increase the order of the controller to take into account the resonant frequency and anti-resonant frequency; Step 5-3: Repeat the process from step 1 to step 4 to improve the control algorithm.
Citation Information
Cited By
Decoupling control method for flexible ball screw system of machine tool
CN122063917A
