JP2003235280A - Motor control device - Google Patents

Abstract

(57) [Summary] [Object] To provide a motor control device that can obtain a good control characteristic by preventing a control delay of a steady-state offset time by a target value filter. An effective command obtained by separating PI and D (P: proportional, I: integral, D: derivative) adjustment operator operations and performing compensation calculation from a control amount from a control target and a command value of the control amount In a motor control device that performs a PI / D adjustment calculation so that a deviation from the value becomes zero and applies the obtained adjustment signal as an operation signal to the control target, a PI adjustment calculation unit and a D A compensation operation unit for obtaining an effective command value for the adjustment operation unit is provided. The compensation operation unit for the PI adjustment operation unit is provided with a second order differentiator for the command value and a first-order low-pass filter in parallel with the control value command value. The output of the second-order differentiator plus the first-order low-pass filter is added to the command value of the amount, and the compensation calculation unit for the D adjustment calculation unit is configured to use the command value of the control amount.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a control device (for example, an NC device such as a machine tool) for making a controlled object follow a temporal change of a command value, and particularly a motor control device for improving disturbance response and followability. Regarding

[0002]

2. Description of the Related Art Conventionally, a general PID as shown in FIG.
In the control, if the PID parameter is set so that the disturbance response is emphasized and the response is improved, the overshoot at the inflection point of the follow-up control becomes large. On the contrary, when the PID parameter is set so that the response to the command value r is emphasized and the response is improved, the disturbance response becomes worse.

As a control method for improving the weaknesses of general PID control, there is a two-degree-of-freedom PID control method in which control is performed with both a PID parameter that improves the disturbance response and a PID parameter that suppresses overshoot. Among the two-degree-of-freedom PID control methods, a target value filter type two-degree-of-freedom P
The ID control has a feature that it can be incorporated into an existing general PID control relatively easily, and is an easy method to realize.

FIG. 6 shows an example of the structure of a conventional target value filter type two-degree-of-freedom PID control (JP-A-6-30472).
No. 6 publication). In the control device shown in the figure, a target value filter 20 is provided in front of the PID calculator 23 that outputs the manipulated variable U to the controlled object 8. Target value filter 2
For 0, the control constant is set so as to be in an optimum state to follow the command value r. When the command value r is input, the reference R compensated by the target value filter 20 is output.

In the PID calculator 23, the control constants are set so as to be in the optimum state for the disturbance response. The deviation that is the difference between the output of the reference R and the output of the controlled object 8 is input to the PID calculator 23, and the manipulated variable U is output.

A manipulated variable U from the PID calculator 23 and a disturbance D are input to the controlled object 8.
Although various calculation methods have been proposed as the target value filter 20, an example of the calculation formula F (s) of the target value filter 20 is shown in (1) below.

F (s) = (1 + α · Ti · s) / (1 + Ti · s) + (− β / (1 + Ti · s) + γ · Ti · s / (1η · Ti · s) × (Ti · s / (1 + Ti · s) (1) In this control device, the target value filter 20 is composed of a proportional equivalent element 16, an integral equivalent element 17, a differential equivalent element 18, and a micro product common term equivalent element 19.

The command value r is first input to the proportional equivalent element 16, the integral equivalent element 17, and the differential equivalent element 18, and the outputs from the integral equivalent element 17 and the differential equivalent element 18 are input to the micro product common term equivalent element 19. To be done. And proportional equivalent element 1
The outputs of 6 and the element corresponding to the microscopic product common term 19 are added and output as the reference R.

On the other hand, the output from the controlled object 8 is input to the differentiating element 22 of the PID calculator. Further, the difference between the output from the controlled object 8 and the reference R from the target value filter 20 is input to the proportional integration element 21.

Then, the proportional-integral element 21 and the derivative element 22
Is output to the controlled object 8 as a manipulated variable U. As parameters of the conventional target value filter type two-degree-of-freedom PID control system configured and operated as described above, a PID parameter in the PID calculator 23 and a proportional equivalent element 16, an integral equivalent element 17, and a derivative of the target value filter 20 are differentiated. Corresponding element 1
8 and target value parameters α, β, γ
There is. In the target value filter type two-degree-of-freedom PID control method, the PID parameter and the target value parameter are adjusted,
We try to find the best response for the process.

The target value filter type two-degree-of-freedom PID control is
This is an effective control method in the case of a step response in which the controlled object is instantaneously positioned at a certain fixed value. FIG. 7 shows an example of step response of a conventional target value filter type two-degree-of-freedom PID control. When the stepped command value is input as shown by 24 (broken line), the step response of the general PID control as shown in FIG. 5 becomes as shown by 25 (dotted line), the overshoot is large, and the positioning time is long. . The step response of the target value filter type two-degree-of-freedom PID control is 26 (solid line).
The overshoot is small and the positioning time is short. Normally, in the target value filter type two-degree-of-freedom PID control, disturbance can be suppressed by the PID control unit and followability can be improved by the target value filter, so a servo system control system excellent in disturbance response and followability can do.

However, the target value filter type 2 described above is used.
The degree of freedom PID control is not always effective in the case of a ramp response in which the command value r changes with time with a certain gradient.

FIG. 8 shows an example of a ramp response of the conventional target value filter type two-degree-of-freedom PID control. When the ramp-shaped command value r is input as shown by 27 (broken line), the ramp response of the target value filter type two-degree-of-freedom PID control is 28 (solid line).
Thus, a time delay such as 29 occurs from the command value in the ramp section which rises with a certain gradient.

A PV switching type two-degree-of-freedom PID control device is known as a control device for improving the weakness of the target value filter type two-degree-of-freedom PID control having such a steady offset (see Japanese Patent Laid-Open No. 7-104807). FIG. 9 shows a structural example of a PV switching type two-degree-of-freedom PID control device. For the target value filter type two-degree-of-freedom PID control, a changeover unit 33 as software means and a changeover switch 30 driven by this changeover unit 33 are newly provided. Then, the switching unit 33
Is provided with a switching point calculating section 32 and a comparator 31 for comparing the output of the switching point calculating section 32 with the output of the controlled object to switch the changeover switch 30.

The PV switching type two-degree-of-freedom PID control device performs general PID control on a ramp section having a constant gradient, and a steady offset time is provided before the switching from the ramp section to a soak section where the gradient becomes zero. When the time point of is exceeded, the changeover switch 30 is switched so as to perform the target value filter type two degrees of freedom PID control.

By using the PV switching type two-degree-of-freedom PID control device, it is possible to eliminate the control delay in the ramp section, reduce overshoot, and perform strong control against disturbance.

[0017]

However, even with the above-mentioned PV switching type two-degree-of-freedom PID control device,
It cannot be said to be effective in the case of a curve response in which the command value r changes with time at an arbitrary gradient. When a curved command value r in which the switching point of the switch is not clear, such as when switching between the ramp section and the soak section shown in FIG. 8, is input, the switch cannot be switched.

Further, in the case of a curve response, since a tracking error (overshoot) is always likely to occur, the target value filter must always be turned on, and the control delay cannot be eliminated. The conventional target value filter type two-degree-of-freedom PID control and PV switching type degree-of-freedom PID control cannot be applied to a processing machine that inputs a curve that changes with time as a command value, and the shape accuracy deteriorates due to control delay. .

The present invention has been made in view of the above problems, and an object of the present invention is to provide a motor control device capable of preventing a control delay of a steady offset time by a target value filter and obtaining good control characteristics. To do.

[0020]

In order to achieve the above object, the present invention separates the PI and D (P: proportional, I: integral, D: derivative) adjustment operator operations and controls from the controlled object. A motor for performing PI / D adjustment calculation so that the deviation from the effective command value obtained by performing compensation calculation from the command value of this control amount becomes zero, and applying the obtained adjustment signal to the control target as an operation signal. The control device is provided with a compensation calculation unit that obtains an effective command value for the PI adjustment calculation unit and the D adjustment calculation unit from the control value command value, and the compensation calculation unit for the PI adjustment calculation unit sets the control value command value to the control value command value. A second order differentiator + 1 order low pass filter for the command value is provided in parallel, and the output of the second order differentiator + 1 order low pass filter is added to the command value for the control amount. The compensation calculation unit for the D adjustment calculation unit is a control amount. Command value Characterized by configured to use.

[0021]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments of the present invention will be described below with reference to the accompanying drawings.

FIG. 1 shows a block diagram of motor control according to the present invention.

A controlled object 8 desired to follow the command value r, and a PID calculation section 7 for feedback controlling the output of the controlled object 8 (comprising a proportional component 4, an integral component 5, and a differential component 6)
And a target value filter 3 for compensating the command value and improving the followability.

Generally, the controlled object has a tracking error (overshoot) in a portion where the speed of the command value r is drastically changed.
Is more likely to occur. Therefore, the target value filter 3 compensates the command value r so that a tracking error does not occur. The part where the speed change of the command value r is large is the part where the acceleration is large, so the second order differentiator 1 calculates the acceleration of the command value. The compensation amount is determined by K of the primary filter 2, and the attenuation time of the compensation amount is determined by T.

The deviation from the output of the controlled object is calculated by adding the compensation amount by the second-order differentiator 1 and the primary filter 2 to the command value r, and the proportional (P) component 4 and integral (I) component 5 of the PID control are calculated.
To enter. Since the compensation amount by the second-order differentiator 1 and the primary filter 2 is provided in parallel with the command value r, by setting K of the primary filter 2 to zero, a general PID as shown in FIG. It can also have the same configuration as the control.

When the command value r which changes stepwise or sharply with time is passed through the second differentiator 1, the compensation amount includes a sharp pulse signal. It is not desirable to move the controlled object in accordance with the pulse-shaped signal included in the compensation amount, as this causes the controlled object 8 to vibrate. Therefore, the deviation between the command value r and the output of the controlled object 8 that is not compensated for in the differential (D) component 6 of the PID control is directly input to the differential (D) component 6 as it is. By doing so, even if the target value filter 3 compensates for the pulse shape, the controlled object 8 can follow the command value r without excessively reacting.

Since the target value filter of the present invention acts only on a portion where the speed of the command value r changes drastically, a control delay occurs in the follow-up characteristic even if the command value r includes a ramp section or a curve section with a certain gradient. do not do.

For example, consider a case where a spindle shaft is rotated at 3800 rpm and a sinusoidal curve having a frequency of 20 Hz is machined in an NC device. The NC device is composed of one spindle shaft and many positioning control shafts. When the spindle shaft rotates 3800 rpm, a periodic disturbance of 63.3 Hz (3800 rpm) occurs on the positioning shaft due to the influence of the rotation.

FIG. 2 shows an example of the effect of the present invention on the followability and the disturbance response. When such processing is performed using general PID control, the positioning tracking deviation becomes 10 (broken line), and due to the influence of disturbance, it is 63.3 Hz (3800 r).
A large processing error occurs in the cycle (pm). Also, 20
A machining error due to the influence of the command value of Hz also occurs as indicated by 11 (dotted line).

When such processing is performed using the control method of the present invention, the positioning follow-up deviation becomes 9 (solid line), which is 63.3 Hz (3800 rpm) due to the influence of disturbance.
Both the machining error of 1 and the machining error due to the influence of the command value of 20 Hz can be reduced to about half of those of the conventional general PID control.

The general PID control as shown in FIG. 5 must improve the followability and the disturbance response by itself, but the control method of the present invention is the conventional target value filter type 2 free type as shown in FIG. As in the case of the degree PID control, the disturbance responsiveness can be increased by the PID control, and the followability can be increased by the target value filter.

FIG. 3 shows an example of the effect of disturbance suppression by the PID set value. For example, the disturbance suppression characteristic in general PID control is as shown by 12 (broken line), and a case where it is particularly vulnerable to a periodic disturbance near 35 Hz is considered. When the control method of the present invention is applied, it is possible to set the PID parameter with emphasis on the disturbance response as in the conventional target value filter type two-degree-of-freedom PID control. The disturbance suppression characteristic at that time is as shown by 13 (solid line), and it is possible to suppress the periodic disturbance near 35 Hz, which is weak in the conventional PID. In this case, 10 Hz to 113
Although it is possible to suppress the periodic disturbance of Hz, the disturbance suppression characteristic 12 (broken line) using the conventional PID set value is better for other frequencies.

Since the target value filter of the present invention is completely independent of the PID control system in the subsequent stage, it is possible to switch and use the PID set value depending on the frequency of the periodic disturbance. For example, after switching to PID parameter setting that emphasizes the disturbance response depending on the frequency of the periodic disturbance, the tracking error can be set to be small with the parameters K and T of the target value filter. FIG. 4 shows an example of the effect of the present invention on the followability. For example, when a sine wave command value is input and the controlled object is made to follow, the tracking error when using general PID control increases significantly as shown by 14 (broken line) as the frequency becomes higher.

However, when the control method of the present invention is applied,
The tracking error is as shown by 15 (solid line), and can be made very small compared with the case of general PID control especially in high frequency tracking. By using the control method of the present invention, the tracking error can be significantly reduced, and the durability in high frequency sinusoidal tracking can be significantly improved.

The target value filter and the PI / D adjustment calculator of the present invention are composed of software, and can be processed by a CPU. An example of an arithmetic expression of the manipulated variable U by the target value filter 3 and the PI / D adjustment arithmetic unit 7 of the present invention is shown in the following equation (2).

U = [{1 + s ² · K / (1 / T + s)} − X] · (P + I / s) + (r−X) · D · s (2) Equation (2) is calculated by the CPU. By inputting the operation amount U to the control target, the same effect as that of the above-described embodiment can be obtained.

[0037]

As is apparent from the above description, according to the present invention, the PI and D (P: proportional, I: integral, D: derivative) adjustment operator operations are separated and the control amount from the controlled object is separated. A motor for performing PI / D adjustment calculation so that the deviation from the effective command value obtained by performing compensation calculation from the command value of this control amount becomes zero, and applying the obtained adjustment signal to the control target as an operation signal. In the control device, from the command value of the control amount, P
A compensation calculation unit for obtaining an effective command value for each of the I adjustment calculation unit and the D adjustment calculation unit is provided, and the compensation calculation unit for the PI adjustment calculation unit parallels the command value of the control amount to the second-order differentiator of the command value + 1 order. Since a low-pass filter is provided and the output of the second-order differentiator + first-order low-pass filter is added to the command value of the control amount, and the compensation calculation unit for the D adjustment calculation unit is configured to use the command value of the control amount, Good control characteristics can be obtained by preventing the control delay of the steady offset time by the target value filter.

[Brief description of drawings]

FIG. 1 is a block diagram of motor control according to the present invention.

FIG. 2 is a diagram showing an example of the effect of the present invention on the followability and the disturbance response.

FIG. 3 is a diagram showing an example of an effect of disturbance suppression by a PID set value.

FIG. 4 is a diagram showing an example of the effect of the present invention on the followability.

FIG. 5 is a diagram showing a structure of general PID control.

FIG. 6 is a conventional target value filter type two-degree-of-freedom PID.
It is a figure which shows the structural example of control.

FIG. 7 is a conventional target value filter type two-degree-of-freedom PID.
It is a figure which shows the example of a step response of control.

FIG. 8 is a conventional target value filter type two-degree-of-freedom PID.
It is a figure which shows the ramp response example of control.

FIG. 9 is a diagram showing a structural example of a PV switching type two-degree-of-freedom PID control device.

[Explanation of symbols]

1 Second-order Differentiator 2 Primary Filter 3 Target Value Filter 4 Proportional Component 5 of PID Control 5 Integral Component of PID Control 6 Differential Component of PID Control 7 PID Calculator 8 Control Target 9 Positioning Tracking Deviation 10 General PID Control Applied Positioning tracking deviation 11 Machining error due to influence of command value 12 Disturbance suppression characteristic in general PID control 13 Disturbance suppression characteristic when PID parameter is set 14 Tracking error when general PID control is used 15 Present invention Error when applying the control method of 16 16 Proportional equivalent element 17 Integral equivalent element 18 Differential equivalent element 19 Small product common term equivalent element

Claims (4)

[Claims] 1. PI and D (P: Proportional, I: Integral, D: Derivative) adjustment operator operations are separated and obtained by performing a compensation operation from a controlled variable from a controlled object and a command value of this controlled variable. In a motor control device that performs a PI / D adjustment calculation so that a deviation from an effective command value becomes zero and applies the obtained adjustment signal as an operation signal to the control target, a PI adjustment calculation unit from the command value of the control amount. And a compensation calculation unit for obtaining effective command values for the D adjustment calculation unit,
The compensation calculation unit for the PI adjustment calculation unit is provided with a second-order differentiator + first-order low-pass filter of the command value in parallel for the command value of the control amount, and outputs the output of the second-order differentiator + first-order low-pass filter to the command value of the control amount. The motor control device is configured to perform addition, and the compensation calculation unit for the D adjustment calculation unit is configured to use the command value of the control amount. 2. By separating the output of the second-order differentiator + first-order low-pass filter of the PI adjustment operation unit, it can be used as a general PID adjustment operation unit that does not perform compensation operation. The motor control device according to item 1. 3. When there is a periodic disturbance, the PID parameter can be switched to a value determined in advance by the disturbance frequency, and the parameter of the second-order differentiator + first-order low-pass filter can be adjusted according to the PID parameter. The motor control device according to claim 1, wherein the motor control device can be switched. 4. The motor control device according to claim 1, wherein the second-order differentiator + first-order low-pass filter and the PI / D adjustment calculation are configured by software and can be processed by a CPU.