Motor position nonlinear disturbance static feedforward compensation method based on system model

Abstract

The invention discloses a motor position nonlinear disturbance static feedforward compensation method based on a system model. The method comprises the following steps: selecting a sampling period T and a closed-loop rotating speed v suitable for a system according to frequency characteristics of nonlinear disturbance in a motor; performing an open-loop experiment on the motor servo control system to obtain a discrete transfer function from a control signal to system open-loop output, and establishing a motor system model; the method comprises the steps that a motor is subjected to closed-loop operation at a low rotating speed, rotating speed, current and angle information during steady-state operation is collected, the rotating speed and the current are substituted into a discrete transfer function, an equivalent current value of nonlinear disturbance at each angle is obtained, and then the equivalent current value is negated to obtain a corresponding feedforward compensation value at each angle; carrying out filtering and weighted average on the obtained equivalent feed-forward compensation values to obtain a group of compensation value sequences with fixed compensation point intervals; and the control unit looks up a table according to an angle signal of the encoder to obtain a feed-forward compensation value of a current angle, and superposes the feed-forward compensation quantity on the control quantity at the output end of the control unit, so that feed-forward compensation for nonlinear disturbance is realized.

Description

Technical Field

[0001] The invention relates to the technical field of servo control, and in particular to a static feedforward compensation method for nonlinear disturbance of motor position based on a system model. Background Art

[0002] Servo motors are widely used in robotic arms, high-precision machine tools and other fields, becoming key components for these devices to achieve precise control and efficient operation. In robotic arms, servo motors achieve complex operating tasks and improve the level of production automation by precisely controlling the movement of each joint; in high-precision machine tools, servo motors ensure the size and shape accuracy of processed parts through precise positioning and stable operation. The accuracy of servo motors directly affects the operating accuracy and repeatability of robotic arms, and determines its reliability and efficiency in automated production; at the same time, in high-precision machine tools, the accuracy of servo motors determines the surface quality and processing accuracy of processed parts, which has an important impact on the product quality and production efficiency of the manufacturing industry. However, in actual operation, servo motors are affected by various nonlinear disturbances, resulting in a decrease in their performance. These nonlinear disturbances are usually related to factors such as friction and cogging force of the motor. Changes in friction can cause instability and decreased precision of motor movement, while cogging force can cause periodic torque fluctuations during motor operation, further affecting the accuracy of motion control. These disturbances not only reduce the operating accuracy and repeatability of robotic arms, but also affect the processing quality and efficiency of high-precision machine tools. Therefore, identifying and compensating these nonlinear disturbances is an important research direction to improve the overall performance of servo motor systems.

[0003] The nonlinear disturbance of the servo motor mainly includes cogging force and friction force. Common methods for suppressing cogging force include structural optimization and feedback control. Structural optimization such as stator skew slot design can weaken the cogging response, but the cost is high and the effect is limited; feedback controllers such as H∞ controllers can improve the disturbance suppression capability and reduce the tracking error in the feed drive, but the system bandwidth requirements are high. There are two main ideas for suppressing friction. The first is to ignore the specific model of friction, regard it as an external disturbance, and suppress it through feedback control methods. This method has a certain suppression effect but poor adaptability. The second is model-based feedforward compensation, which predicts and compensates friction by accurately analyzing the friction model. The dynamic Lugre model can provide better compensation effect, but the model identification is complex, there are many parameters, and it is difficult to apply in practice. The static Stribeck model is simple and easy to use, but the compensation effect is relatively poor. Summary of the invention

[0004] The purpose of the present invention is to overcome the above-mentioned shortcomings and deficiencies of the prior art and to provide a static feedforward compensation method for motor position-related nonlinear disturbances based on a system model.

[0005] The present invention establishes a discrete transfer function model of the motor system, substitutes the speed and current of the motor in stable operation into the transfer function model to calculate the equivalent feedforward compensation value of the nonlinear disturbance at each position. The method has a small amount of calculation, is easy to implement, has high compensation accuracy, and is strong in robustness.

[0006] The present invention is achieved through the following technical solutions:

[0007] A method for static feedforward compensation of motor position-related nonlinear disturbance based on a system model comprises the following steps:

[0008] S1. Construct a motor servo control system, which includes a control unit, a controlled object, an actuator, and a feedback measurement device, and collects system output signals and control signals when the motor servo control system is running in a closed loop;

[0009] S2. According to the frequency characteristics of the nonlinear disturbance in the motor, select the sampling period T and closed-loop speed v suitable for the system;

[0010] S3. Conduct an open-loop experiment on the motor servo control system, obtain the discrete transfer function from the control signal to the system open-loop output, and establish a motor system model;

[0011] S4. The motor is placed in closed-loop operation at a speed v, and the speed, current, and angle information during steady-state operation are collected. The speed and current are substituted into the discrete transfer function to obtain the equivalent current value of the nonlinear disturbance at each angle, and then the equivalent current value is inverted to obtain the corresponding feedforward compensation value at each angle;

[0012] S5. Filtering the obtained equivalent feedforward compensation value and performing weighted averaging to obtain a set of compensation value sequences with fixed compensation point intervals;

[0013] S6. The control unit obtains the feedforward compensation value of the current angle by looking up the table according to the encoder angle signal. At the output end of the control unit, the control amount is superimposed on the feedforward compensation amount to achieve feedforward compensation for nonlinear disturbances.

[0014] Further, in step S1, the controlled object is a controlled motor, the actuator is a driver of the motor, and the feedback measurement device is a servo motor encoder on the motor;

[0015] The control unit is used to set the command signal, collect sensor data, calculate and send the control rate, and the servo motor encoder is used to accurately measure and feedback the actual position and speed of the motor.

[0016] Furthermore, the control unit includes a feedback controller and a feedforward controller. The feedback controller adopts a PI controller. The feedback controller outputs a feedback control rate to the driver to generate a corresponding current signal according to the deviation between the motor command signal and the servo motor encoder measurement signal; the feedforward controller obtains a feedforward compensation value of the current angle by looking up the table according to the encoder angle value, which is used to compensate for the position-related deviation of the motor speed caused by nonlinear disturbance.

[0017] Furthermore, in step S2, in order for the collected data to correctly describe the characteristics of the nonlinear disturbance, according to the sampling theorem, the sampling frequency must be at least twice the highest frequency of the signal. In engineering, it may even need to be ten times higher to be well reproduced. The nonlinear disturbance factor is position, and the factors affecting the sampling frequency of position are the closed-loop speed and sampling period of the motor. Therefore, the sampling period and speed of the system are selected according to the characteristics of the nonlinear disturbance. Therefore, we first need to select the sampling period and closed-loop speed of the system according to the characteristics of the nonlinear disturbance.

[0018] Furthermore, we need to select the sampling period and closed-loop speed according to the frequency characteristics of the nonlinear disturbance before we can proceed to the next step. Specifically, since this method is aimed at position-related disturbances, we can try to give the system a stable step input. After obtaining the system's response, we can use fast Fourier transform, short-time Fourier transform and other methods to view the frequency characteristics of the disturbance signal. In addition to this method, we can also use the mechanism method to analyze the frequency characteristics of the position-related nonlinear disturbances in the system.

[0019] Furthermore, after obtaining the frequency characteristics of the disturbance signal, since our system object is a motor and the nonlinear disturbance is related to the position, the system sampling period can be calculated by the following formula:

[0020]

[0021] Where T is the sampling period of the system, v is the target speed when the system is running at a stable low speed, k is the cutoff frequency magnification factor, which is preferably 10 or above. If this requirement cannot be met, it must be at least 2. max is the highest frequency of the nonlinear disturbance of the system.

[0022] Furthermore, in step S3, an open-loop experiment is performed on the motor servo control system, a pseudo-random signal is used as an excitation signal to control the driver to generate a corresponding current signal, the motor speed change caused by the current excitation is measured by the encoder, and a discrete transfer function model of the motor system is established by using a system identification method. The discrete transfer function of the motor system is expressed as:

[0023]

[0024] Where I(z) is the Z-transform of the motor control current i(k), V(z) is the Z-transform of the motor shaft encoder velocity output v(k) caused by the current i(k) input, G(z) is the discrete transfer function of the Z-domain motor model, and k represents the kth sampling time.

[0025] Furthermore, in step S4, the nonlinear disturbance is equivalent to the current i d (k), and as part of the control current i(k), that is, i(k) consists of two parts i(k) = i power (k)-i d (k), where i power (k) is the output signal of the controller. Then the discrete transfer function of the motor system is expressed as:

[0026]

[0027] Where G(z) is the discrete transfer function of the Z-domain motor model, V(z) is the Z-transform of the motor speed output v(k) caused by the current i(k) input, and I(z) is the Z-transform of the motor control current i(k). When there is no nonlinear disturbance in the system, i(k) = i power (k). At this time, I(z) is the driver control current i power Z-transform of (k)i power (z). When there is nonlinear disturbance in the system, i(k)=i power (k)-i d (k), k represents the kth sampling moment, and the corresponding Z transformation is: I(z) = I power (z)-I d (z).

[0028] Furthermore, according to the general mechanism model of the motor, we can know that when the motor current is given and the motor speed is output, the model order of the motor system is second order, so the discrete transfer function of the motor system can be expressed as:

[0029]

[0030] Expanding the above equation into a difference equation form, it can be expressed as:

[0031] v(k)+a1v(k-1)+a2v(k-2)=b1[i power (k-1)-i d (k-1)]+b2[i power (k-2)-i d (k-2)];

[0032] In the above formula, v(k) is the speed value of the motor system at the kth sampling period, i power(k) is the motor system driver control current value at the kth sampling cycle, i d (k) is the equivalent current value of the nonlinear disturbance of the motor system at the kth sampling period, and k represents the kth sampling moment.

[0033] Furthermore, since the data we collected is data from N consecutive moments, we can combine the following N-2 equations, expressed as:

[0034]

[0035] In the above formula, v(k) is the speed value of the motor system at the kth sampling period, i power (k) is the motor system driver control current value at the kth sampling cycle, To obtain the equivalent current value of the nonlinear disturbance of the motor system at the kth sampling period, k represents the kth sampling moment.

[0036] Furthermore, in the above equations, v(k), i power (k) is a known variable, is the variable to be determined, and there are N-2 equations and N-1 unknowns in the formula. Therefore, in order to solve the equation smoothly, and because the motor speed is low and the sampling period is short, we will They are approximately equal, so the system of equations can be expressed as:

[0037]

[0038] In the above formula, v(k) is the speed value of the motor system at the kth sampling period, i power (k) is the motor system driver control current value at the kth sampling cycle, To obtain the equivalent current value of the nonlinear disturbance of the motor system at the kth sampling period, k represents the kth sampling moment. At this time, the system has N-1 equations and N-1 unknowns, which can be solved smoothly. The specific solution of the equation is as follows:

[0039]

[0040] Furthermore, in step S5, after obtaining the feedforward compensation value at each sampling point, in order to improve the accuracy of the data, smooth the data, and improve the data quality, a low-pass filter can be used to filter out the high-frequency noise in the data. The specific cutoff frequency should be determined according to the frequency characteristics of the disturbance, and the highest frequency of the nonlinear disturbance can be used as the cutoff frequency.

[0041] Furthermore, after obtaining the filtered compensation value data, in order to improve the efficiency of the table lookup method, we should select the appropriate compensation point interval according to the frequency characteristics of the disturbance. Specifically, in step S2, we have obtained the highest frequency value of the system nonlinear disturbance. According to the sampling theorem, if we want to accurately reproduce a signal, its sampling frequency must be at least twice the highest frequency of the signal. In engineering, it even needs to be ten times to be well reproduced. Therefore, if we want to reproduce the characteristics of nonlinear disturbance well, the interval of the sampling points should also meet this condition. Therefore, after obtaining the frequency characteristics of the system, we can set the compensation point interval according to the following formula.

[0042]

[0043] Where Δ is the compensation point interval, k is the cutoff frequency magnification, preferably 10 or more, if this requirement cannot be met, at least 2 or more is required, f max is the highest frequency of the nonlinear disturbance of the system.

[0044] Furthermore, after obtaining the appropriate compensation point interval, we set the appropriate compensation interval according to the system characteristics. For example, the normal servo motor handles the position-related disturbance compensation interval of 0°-360°. After obtaining the compensation interval, the compensation value can be obtained. At each compensation point, the compensation value data before and after the compensation point are obtained by linear interpolation. After traversing, a set of compensation value sequences with fixed compensation point intervals can be obtained.

[0045] Furthermore, after obtaining the compensation value sequence with a fixed compensation point interval, if the compensation value sequence contains multiple groups of compensation values ​​in the same compensation interval, we can average all compensation values ​​of the same compensation point to improve the accuracy of the data, smooth data fluctuations, and obtain the final compensation value sequence.

[0046] Further, in step S6, the feedforward controller receives the encoder angle value as input, looks up the table to obtain the compensation control amount corresponding to the current angle value, superimposes the compensation control amount with the feedback control amount output by the feedback controller and the equivalent current of the nonlinear disturbance, and inputs the superimposed control amount to the driver of the servo motor to realize feedforward compensation. The superimposed control amount is expressed as:

[0047]

[0048] In the formula, i power (k) is the feedback control quantity output by the feedback controller during the kth sampling period; i d (k) is the equivalent current value of the nonlinear disturbance of the motor system at the kth sampling period; is the compensation control quantity output by the feedforward controller at the kth sampling period, and i(k) is the actual input control quantity of the motor at the kth sampling period.

[0049] Furthermore, if the accuracy of a single compensation cannot meet the requirement, steps S3-S6 may be repeated after the compensation value is input to compensate for the nonlinear disturbance multiple times to improve the compensation accuracy.

[0050] Compared with the prior art, the present invention has the following advantages and effects:

[0051] The sensor used in the present invention is only the encoder of the servo motor, and the theory is simple. The solution is easy to implement and the cost is not high.

[0052] The present invention analyzes the influence of position-related nonlinear disturbance on the operation of the motor, and derives a method for calculating the equivalent compensation value of the motor nonlinear disturbance based on the motor mechanism model, thereby effectively reducing the speed fluctuation during the steady-state operation of the motor.

[0053] Compared with the existing servo motor control system method of achieving compensation through feedback control, the present invention is based on system model design and uses a table lookup method to directly obtain the feedforward compensation value of the current position, which has a faster compensation speed for nonlinear disturbances related to the position. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] Figure 1 It is a flow chart of the motor nonlinear disturbance feedforward compensation method based on the system model of the present invention.

[0055] Figure 2 This is the system block diagram after adding the feedforward controller.

[0056] Figure 3 This is the timing diagram of the motor compensation value obtained by preliminary calculation.

[0057] Figure 4 This is the final motor compensation value waveform after the original compensation value is filtered and weighted averaged.

[0058] Figure 5 Comparison of system tracking errors with and without a feedforward controller. DETAILED DESCRIPTION

[0059] The present invention is further described in detail below in conjunction with specific embodiments.

[0060] See also Figure 1 The present invention discloses a method for static feedforward compensation of motor position-related nonlinear disturbance based on a system model, comprising the following steps:

[0061] S1. Construct a motor servo control system, which includes a control unit, a controlled object, an actuator, and a feedback measurement device, and collects system output signals and control signals when the motor servo control system is running in a closed loop;

[0062] S2. According to the frequency characteristics of the nonlinear disturbance in the motor, select the sampling period T and closed-loop speed v suitable for the system;

[0063] S3. Conduct an open-loop experiment on the motor servo control system, obtain the discrete transfer function from the control signal to the system open-loop output, and establish a motor system model;

[0064] S4. Place the motor in closed-loop operation at a speed v, collect the speed, current, and angle information during steady-state operation, substitute the speed and current into the discrete transfer function, obtain the equivalent current value of the nonlinear disturbance at each angle, and then invert the equivalent current value to obtain the corresponding feedforward compensation value at each angle.

[0065] S5. Filter the obtained equivalent feedforward compensation value and perform weighted averaging to obtain a set of compensation value sequences with fixed compensation point intervals.

[0066] S6. The control unit obtains the feedforward compensation value of the current angle by looking up the table according to the encoder angle signal. At the output end of the control unit, the control amount is superimposed on the feedforward compensation amount to achieve feedforward compensation for nonlinear disturbances.

[0067] Since motors are generally mounted on specific equipment as drive devices, this example uses a gimbal pitch axis that uses a motor as a drive device and has undergone preliminary gravity compensation as an example to illustrate the static feedforward compensation method for motor position-related nonlinear disturbances based on a system model.

[0068] Specifically, see Figure 2 In step S1, the control unit includes a feedback controller and a feedforward controller, which are used to receive control instructions, collect sensor data and calculate control rate. The controlled object is the pitch axis of the pan-tilt platform, and the actuator is the servo motor of the motor pitch axis. The feedback measurement device is the encoder of the servo motor, which is used to accurately measure and feedback the actual position of the motor pitch axis. The implementation object of this example is the pan-tilt platform pitch axis motor, so the data of the motor pitch axis motor encoder is used to represent the rotation speed of the pitch axis.

[0069] The feedback control in the control unit adopts a PI controller. The feedback controller outputs a feedback control rate to the driver to generate a corresponding current signal based on the deviation between the motor command signal and the servo motor encoder measurement signal. The feedforward controller obtains the feedforward compensation value of the current position by looking up the table according to the encoder angle value, and transmits the feedforward compensation value to the motor driver to compensate for the position-related deviation of the motor speed caused by nonlinear disturbances.

[0070] In step S2, according to the mechanism method, we know that the nonlinear disturbance with the highest disturbance frequency in the system is the cogging force disturbance, and the frequency of this disturbance is 24 times / turn. Through actual measurement, it is determined that the minimum speed at which the system can operate stably is 100° / s, so the sampling period of the system is calculated as follows:

[0071]

[0072] However, since the minimum sampling period for the system we selected to operate stably is 0.02s, and the cutoff frequency magnification factor is greater than 2 when the system is 0.02s, 0.02s is selected as the sampling period of the system.

[0073] In step S3, an open-loop modeling experiment is performed on the pan-tilt servo control system. In this embodiment, a random signal narrow square wave instruction is applied to the pitch axis motor in the open loop of the system, and the angle data output by the motor encoder is measured. The discrete transfer function of the pan-tilt servo system with current as a given and speed as an output is obtained by modeling using the system identification method:

[0074]

[0075] Wherein, z is the time shift operator in the Z domain; P(z) is the discrete transfer function of the Z domain pan / tilt system, and the system sampling period of this embodiment is T=0.02s.

[0076] In step S4, the collected speed and current are substituted into the discrete transfer function to obtain the equivalent feedforward compensation value at each angle. The current compensation value at each moment without filtering can be found in Figure 3 .

[0077] In step S5, a series of compensation value data collected is subjected to zero-phase filtering. According to the mechanism method, we know that the nonlinear disturbance with the highest disturbance frequency in the system is the cogging force disturbance, and the frequency of this disturbance is 24 times / turn. In order to further improve the compensation effect, we increase the cutoff frequency magnification to 25, so the compensation point interval of the system is calculated as follows:

[0078]

[0079] Then, 0° is used as the starting point of the compensation value, and 0.6° is used as the compensation point interval. At each compensation point, the compensation value data before and after the compensation point are linearly interpolated to obtain the compensation value of the compensation point, and the compensation value sequence of the entire motor motion range is obtained; then the compensation value data is divided into 360° as a cycle, and the average value of all compensation values ​​at the same compensation point is calculated. For the specific compensation value sequence, please refer to Figure 4 .

[0080] In step S6, the obtained compensation value sequence is added to the feedforward compensation, the system is set in the closed-loop compensation state, and the speed setting of the system before and after compensation is set to 100° / s. The effect of feedforward compensation is compared. For specific waveform diagrams, please refer to Figure 5 ,According to the experimental results, it can be seen that the feedforward compensation method used in this ,experiment can effectively compensate for nonlinear disturbances.

[0081] As described above, the present invention can be better implemented.

[0082] The implementation methods of the present invention are not limited to the above-mentioned embodiments, and any other changes, modifications, substitutions, combinations, and simplifications made without departing from the spirit and principles of the present invention shall be equivalent replacement methods and shall be included in the protection scope of the present invention.

Claims

1. A static feedforward compensation method for nonlinear disturbance of motor position based on system model, characterized in that: The following steps are involved: S1. Construct a motor servo control system, which includes a control unit, a controlled object, an actuator, and a feedback measurement device, and collects system output signals and control signals when the motor servo control system is running in a closed loop; S2. According to the frequency characteristics of the nonlinear disturbance in the motor, select the sampling period T and closed-loop speed v of the corresponding system; S3. Conduct an open-loop experiment on the motor servo control system, obtain the discrete transfer function from the control signal to the system open-loop output, and establish a motor system model; S4. The motor is placed in closed-loop operation at a speed v, and the speed, current, and angle information during steady-state operation are collected. The speed and current are substituted into the discrete transfer function to obtain the equivalent current value of the nonlinear disturbance at each angle, and then the equivalent current value is inverted to obtain the corresponding feedforward compensation value at each angle; S5. Filtering the obtained equivalent feedforward compensation value and performing weighted averaging to obtain a set of compensation value sequences with fixed compensation point intervals; S6. The control unit obtains the feedforward compensation value of the current angle by looking up the table according to the encoder angle signal. At the output end of the control unit, the control amount is superimposed on the feedforward compensation amount to achieve feedforward compensation for nonlinear disturbances.

2. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 1, characterized in that: In step S1, the controlled object is the controlled motor; the actuator is the motor driver; the feedback measurement device is the servo motor encoder on the motor; the control unit is used to set the command signal, collect sensor data, calculate and send the control rate; the servo motor encoder is used to accurately measure and feedback the actual position and speed of the motor.

3. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 2, characterized in that: The control unit includes a feedback controller and a feedforward controller; The feedback controller uses a PI controller. The feedback controller outputs a feedback control rate to the driver to generate a corresponding current signal according to the deviation between the motor command signal and the servo motor encoder measurement signal; The feedforward controller obtains the feedforward compensation value of the current angle by looking up the table according to the encoder angle value, which is used to compensate for the position-related deviation of the motor speed caused by nonlinear disturbance.

4. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 1, characterized in that: In step S2, in order for the collected data to correctly describe the characteristics of the nonlinear disturbance, according to the sampling theorem, the sampling frequency is at least twice the highest frequency of the signal, and the nonlinear disturbance factor is the position, and the factors related to the sampling frequency of the position are the closed-loop speed and sampling period of the motor. Therefore, the sampling period and closed-loop speed of the system are selected according to the characteristics of the nonlinear disturbance.

5. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 1, characterized in that: In step S3, an open-loop experiment is performed on the motor servo control system. A pseudo-random signal is used as an excitation signal to control the driver to generate a corresponding current signal. The motor speed change caused by the current excitation is measured by the encoder. A discrete transfer function model of the motor system is established using a system identification method. The discrete transfer function of the motor system is expressed as: In the formula, I(z) is the Z-transform of the motor control current i(k); V(z) is the Z-transform of the motor shaft encoder velocity output v(k) caused by the current i(k) input; G(z) is the discrete transfer function of the Z-domain motor model; k represents the kth sampling time.

6. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 1, characterized in that: In step S4, the nonlinear disturbance is equivalent to the current i d (k), and as part of the control current i(k), that is, i(k) consists of two parts i(k) = i power (k)-i d (k), where i power (k) is the output signal of the controller; then the discrete transfer function of the motor system is expressed as: In the formula, G(z) is the discrete transfer function of the Z-domain motor model; V(z) is the Z-transform of the motor shaft encoder velocity output v(k) caused by the current i(k) input; I(z) is the Z-transform of the motor control current i(k); When there is no nonlinear disturbance in the system, i(k)=i power (k); at this time, I(z) is the driver control current i power Z transform of (k) power (z); When there is nonlinear disturbance in the system, i(k)=i power (k)-i d (k), k represents the kth sampling moment, and the corresponding Z transformation is: I(z) = I power (z)-I d (z).

7. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 1, characterized in that: In step S5, after obtaining the feedforward compensation value at each sampling point, in order to improve the accuracy of the data, smooth the data, and improve the data quality, a low-pass filter is used to filter the high-frequency noise in the data. The specific cutoff frequency should be determined according to the frequency characteristics of the disturbance, with the highest frequency of the nonlinear disturbance as the cutoff frequency.

8. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 1, characterized in that: In step S6, the feedforward controller receives the encoder angle value as input, looks up the table to obtain the compensation control amount corresponding to the current angle value, superimposes the compensation control amount with the feedback control amount output by the feedback controller and the equivalent current of the nonlinear disturbance, and inputs the superimposed control amount into the driver of the servo motor to realize feedforward compensation. The superimposed control amount is expressed as: In the formula, i power (k) is the feedback control quantity output by the feedback controller; i d (k) is the equivalent current value of the nonlinear disturbance of the motor system is the compensation control quantity output by the feedforward controller; i(k) is the actual input control quantity of the motor.

9. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 4, characterized in that: In step S2, for position disturbance, a stable step input can be given to the system. After obtaining the system response, the frequency characteristics of the disturbance signal can be viewed using methods such as fast Fourier transform and short-time Fourier transform, or the frequency characteristics of the position-related nonlinear disturbance existing in the system can be analyzed using the mechanism method; After obtaining the frequency characteristics of the disturbance signal, since the system object is a motor and the nonlinear disturbance is related to the position, the system sampling period can be calculated by the following formula: In the formula, T is the sampling period of the system, v is the target speed when the system is running at a stable low speed. k is the cutoff frequency magnification factor, which should be 10 or above. If this requirement cannot be met, it should be at least 2. max is the highest frequency of the nonlinear disturbance of the system.

10. The method for static feedforward compensation of nonlinear disturbance of motor position based on system model according to claim 8, characterized in that: In step S6, if the accuracy of the single compensation control amount cannot meet the requirements, steps S3-S6 can be repeated after the compensation value is connected to compensate for the nonlinear disturbance multiple times to improve the compensation accuracy.

Images