JPH0531388B2 - - Google Patents

Description

[Detailed Description of the Invention] [Field of Industrial Application] The present invention relates to a motor speed control method and a speed control device thereof, and particularly to a method for detecting the rotational speed of a DC motor and providing feedback (P) proportional to the rotational speed. This research is concerned with a proportional-integral control (PI control) method and its control device, which provides feedback (I) proportional to the integral and prevents irregularities in rotation that occur during start-up characteristics and load fluctuations.

[Prior art] Conventionally, when simply controlling the speed of a DC motor, the rotation speed signal detected by a rotation speed detector attached to the rotation shaft of the DC motor is compared with a speed command signal, and the obtained value is calculated. The error signal thus obtained is appropriately amplified and fed back to the DC motor. The rotational speed of the DC motor is controlled in proportion to the error signal. Furthermore, since the error signal is proportional to the rotational speed, this control is called proportional control (P control, P stands for Proportional).

In the above proportional control, the DC motor is driven by the error signal to obtain a predetermined rotational speed, so there is a problem in that it is not possible to make the rotational speed signal match the speed command signal during operation.

This problem is based on integral control (I control, I is
This can be solved by Integration I).
In other words, if the DC motor is driven by a signal obtained by integrating the error signal, when the error signal becomes zero, the integrated signal will be held at a constant value, and the rotational speed signal will match the speed command signal, so the error in the rotational speed will occur. I don't.

However, since integral control alone tends to reduce the phase margin of the feedback loop and cause the response to become unstable, in practice PI control combined with proportional control has been widely adopted.

However, since the characteristics of the DC motor described above vary greatly depending on factors such as the inertia factor, drive voltage, and load torque that act on the motor, it is necessary to reset the characteristics of the PI control described above in accordance with the magnitude of each change. It was hot.

In order to shorten the transient response time in the above-mentioned PI control, the method disclosed in Japanese Patent Application Laid-Open No. 53-79175 has a method that produces a large error signal when the speed command signal changes in a stepwise manner, such as during startup. During this period, the I control loop was cut off to shorten the rise time, and the I control loop was connected only after the error signal reached a sufficiently small value.

Although this method can shorten the response time, the response may become unstable if the inertia factor of the load changes, so it is necessary to reset the constants of the I control loop when the inertia factor of the load changes. Ta.

In addition, in Japanese Patent Application Laid-Open No. 52-155503, a reduced pass filter is used instead of the above PI control element,
Changes in the inertia factor of the load were detected from the start-up characteristics when the DC motor was started, and the time constant of the above-mentioned reduced pass filter was controlled to stabilize the response characteristics regardless of changes in the inertia factor of the load. .

However, this method has a problem in that an error in the rotational speed occurs because it is not possible to drive the DC motor by making the error signal, which is the difference between the rotational speed of the DC motor and its speed command value, zero.

In addition, this method also includes an outline of the response characteristics at startup.
Although it is possible to stabilize the system, there is a problem in that it is difficult to always accurately maintain the system at a predetermined response characteristic regardless of changes in the inertia factor of the load connected to the motor.

Hereinafter, each of the above-mentioned problems will be theoretically explained to clarify the purpose of the present invention.

FIG. 12 is a block diagram showing the structure of the invention described in the above-mentioned Japanese Patent Application Laid-Open No. 52-155503.

3 is a low-pass filter shown in FIG. 2b of the same specification, and its transfer function is expressed as 1/(1+sτ _F ). However, τ _F is the time constant of the reduced pass filter 3,
s is a Laplacian. Similarly, A ₂ /(1+
sτ _M ) is the transfer function of the motor 5. The time constant τ _M changes in proportion to the inertia factor of the rotor of the motor 5 and the load. A ₁ is the gain of the motor drive circuit. The rotational speed N of the motor 5 is converted into a pulse rate n by a rotational speed detector 6 including a Hall element, and compared with the frequency f _r of the reference oscillator 1 by a frequency comparator 2 .

From the above block diagram, the relationship between the reference frequency _fr and the motor rotational speed N is derived as shown in equation (1).

N/f _r =A/1+AK _N・1/(s/ω _O ) ² +2ζ(s/ω
_O )+1(1) However, A=A ₁ A ₂ , It is.

The above ζ is a damping coefficient. The state of ζ=1/√2 is called critical braking, and at this time, the response of the system is vibration-free and the fastest response speed is obtained, so the design goal is usually to approach this critical braking condition.

According to equation (2), the value of ζ changes according to the time constant τ _M of the motor, so by detecting the change in τ _M and controlling the time constant τ _F of the low-pass filter 3, τ _F / If the value of τ _M is kept constant, the value of ζ will not change from the preset value, so a good response will always be obtained.

[Problems to be Solved by the Invention] However, the method disclosed in JP-A-52-155503 has a problem in that changes in the time constant τ _M of the motor cannot be detected accurately. As a result, the value of ζ cannot be maintained at a predetermined value, and the braking characteristics deviate from the predetermined characteristics. In the prior art described above, τ _M is determined from the value obtained by integrating the rising portion of the step response (indicinal response) of the motor.

When this step response is derived from equation (1), it becomes equation (2).

= (A/1+AK _N ) [1-e ^-at (cosb t+a/bsinbt)] (3) However, a=ω _O ζ b=ω _O √1- ² (4) With a step width of F _r = f _r be.

Equation (5) is obtained by integrating Equation (3) from time 0 to t.

∫ ^t ₀ (N/F _r )dt = t-1/a ² + b ² [e ^-at {-2acosbt + (b-a ² /b) sinbt} + 2a] (5) However, for simplicity, here ( A/(1+
AK _N ) is set to 1.

Extracting the value of τ _M from equation (5) requires a sophisticated arithmetic device. Such a calculation device not only impairs the economic efficiency of the motor control device, but also the delay caused by the calculation time is extremely detrimental to increasing the response speed, and may cause the response to become unstable in some cases.

Also, in order to simplify the calculation of τ _M , for example, if at, bt, etc. are assumed to be minute values, we can approximate them as e ^-at ≒ 1 − at cosbt ≒ 1, sinbt ≒ bt (6). 5) becomes equation (7), ∫ ^t ₀ (N/F _r )dt≒(1−2ζ ² )ζω _O t (7) As expected, the above calculation is not so simple.

As described above, although the above-mentioned conventional technology allows for very approximate control, it is not possible to perform accurate control.

The object of the present invention is to provide a motor speed control method that can accurately detect the specific number τ _M of the motor or the inertia factor acting on its rotating shaft, and also that can follow the speed command value without error. The goal is to provide equipment.

[Means for Solving the Problems] In order to solve the above problems, the present invention includes a motor driven by an inverter, a speed generator that detects the rotational speed of the motor, and a speed signal that converts the output of the speed generator into a speed signal. a speed detection circuit for converting the speed to a speed command value, means for generating a difference signal between the output of the speed detection circuit and a speed command value, and means for generating a proportional term proportional to the error signal and a time integral term of the error signal. , in a motor speed control device that drives the inverter by the sum of the proportional term and the time integral term, a predetermined time width is set when starting the motor or when changing the speed command value. The rotational acceleration of the motor is determined from the variation range of the speed signal within the time width, and when the rotational acceleration is larger than a predetermined value determined in advance based on the load torque and inertia factor of the motor, the coefficient of the time integral term is increased. However, when the rotational acceleration is smaller than the predetermined value, the coefficient of the time integral term is reduced.

[Operation] The motor speed control method and its speed control device of the present invention configured as described above detect the rotational acceleration α of the motor and quickly and accurately calculate the time constant τ _M of the motor using a simple procedure. Since the braking coefficient ζ of the speed control system of the motor is corrected, the response characteristics of the motor speed control system can always be optimally controlled in response to changes in the inertia factor of the load acting on the motor, load torque, etc.

As a result, the response speed of the system at startup or when changing the speed command is increased, and at the same time unnecessary vibrations such as hunting are prevented.

[Embodiment] FIG. 1 is a block diagram illustrating the configuration of the present invention. A speed detector 4 detects the rotational speed N of the motor 2 and converts it into a speed signal N. Speed signal N
is compared with the speed command N _C , and a difference signal between the two, that is, an error (N _C −N), is applied to the motor 2 via the PI circuit 1 , driver 7 , inverter 3 , etc. Furthermore, the rotational acceleration α of the motor 2 is detected by the acceleration detector 5 and the PI circuit 1
is applied to

FIG. 2 is a block diagram representing FIG. 1 using a transfer function. Reference numeral 21 is a block in which the driver 7, the switching circuit 3, and the motor 2 are collectively represented, and its transfer function is expressed as A/(1+sτ _M ).
K _N is a conversion coefficient for the output of the speed detector 4. The PI circuit is composed of a proportional circuit that multiplies the error (N _C -N) by K _P , an integrator circuit for the error voltage (N _C -N), and an adder for both. K _I /s is the transfer relationship of the integrating circuit. Further, sK _C is a transfer function of the acceleration detector 5, and its output V _C controls the value of the above-mentioned integral coefficient K _I. First, find the transmission coefficient N/V _r of the system when K _C is zero, and from this calculate the motor time constant τ _M and integral coefficient K _I necessary to keep the system braking coefficient ζ at a predetermined constant value. After deriving the relationship, it will be explained that the above condition can be satisfied by the acceleration detector 5.

Normally, the value of (K _P K _N A) is sufficiently large compared to 1, so taking this into account, the transfer function N/ in Figure 2 is
N _O becomes as shown in equation (8).

N/N _O =1/K _N・1+2ζ(s/ω _O /(s/ω _O ) ² +2
ζ(s/ω _O )+1(8) However From this, it can be seen that in order to keep the control coefficient ζ constant with K _P , K _N , A, etc. being constant values, it is sufficient to control the product of K _I and τ _M to be a constant value.

In the present invention, the value of τ _M is detected from the rotational acceleration α of the motor 2 during the step response of the system. The response when the motor 2 is started is an example of this step response. Therefore, when the step response of equation (8) is determined, equation (10) is obtained.

1/K _N {1-e ^-at (cosbt-a/bsinbt)} (10) From equation (10), the rotational acceleration α at the time of starting motor 2, that is, at t=0, is determined as shown in equation (11). Become.

α=2a=K _N AK _P /ζ _M (11) From this, since K _P , K _N , and A are constants, the time constant τ _M of the motor can be accurately determined by detecting the rotational acceleration α. Furthermore, since the procedure for calculating τ _M from α is simple, the computation time that is harmful to control is short and does not pose a practical problem.

Furthermore, since the control coefficient ζ is given by Equation (9), it is sufficient to find the above τ _M and control the value of K _I so that τ _M K _I becomes a constant value.

FIG. 3 is an example of an actual circuit of the block diagram shown in FIG. A load 21 is connected to the motor 2, and the time constant τ _M is determined by the total value of the inertia factors of the load 21 and the motor 2. Further, the speed detector 4 is constituted by a speed generator 41 coupled to the rotating shaft of the motor 2 and a speed detection circuit 42.
1 generates a predetermined number of pulses per revolution of the motor 2, and the speed detection circuit 42 generates a time rate signal of the number of pulses, that is, a speed signal n, and applies this to the microcomputer 6.

The microcomputer 6 and the acceleration detector 5
It has the function of PI circuit 1, and calculates the error voltage (N _C ) based on the speed signal n and speed command signal N _C.
-N), calculate and add a proportional term proportional to this and an integral term obtained by integrating this, and then use the above added value to generate a pulse width modulated wave necessary for driving the switching circuit 3. 12 is generated and applied to the driver 7.

Further, the microcomputer 6 calculates the rotational acceleration α of the motor 2 from the signal n according to the activation signal 17, and corrects the value of the integral coefficient K _I of the integral calculation according to the calculated value. That is, when α is too large, K _I is increased, and when α is too small, the value of K _I is determined.

Further, a rotation direction command 16 for the motor 2 is applied to the microcomputer 6, which is applied to the driver 7.
is transmitted to set the rotation direction of the motor 2.

4 and 5 are flowcharts showing the process of calculations performed by the microcomputer 6. FIG.

In FIG. 4, the start signal 1 is activated at step S1.
7 and the rotational direction command 16, the rotational speed signal n is compared with the speed command signal N _r , and step S2
Calculate the above proportional term from the error signal obtained by
In the next step S3, the acceleration α is obtained, and in step S4, the integral term is calculated. next step
In S5, the duty of the pulse width modulated wave 12 is calculated from this integral term.

FIG. 5 is a diagram showing the flow shown in FIG. 4 in more detail.

When the above calculation starts at step S6,
At step S7, the time width t _S1 and the initial value K _d of the above integral are determined.
is set. Since the speed command signal N _r is changed even during system operation, the value of the integral term at the time of this command change is held as the initial value. At step S8, the speed command signal N _C , the rotational speed signal n, and the rotational direction command (R〓) 16 are taken in, and the process proceeds to step S8.
Calculate the proportional term in S9.

If the rotational speed signal n is different from the speed command signal _Nr in the next step S10, it is detected in step S11 that the time t has passed the time width tS1, and the process proceeds to _{step S11} .
Acceleration is calculated in S12. This calculation is based on the difference Δn between the rotational speed signal n between time t=0 and time t=t _S1 .
Find it by dividing by t _S1 .

In step S13, the time constant τ _M of motor 2 is calculated from the above acceleration α according to equation (8), and then the next step is
In S14, according to equation (6), the value of the integral coefficient K _I that compensates for the change in τ _M is determined, and the control dynamic coefficient ζ is maintained at a predetermined value.

In step S15, the switching circuit _duty factor (ON-OFF
ratio) is calculated and output in step S16.

Figures 6 and 7 show the speed detection circuit 42 in Figure 3.
FIG. 4 is a diagram showing an example of the operation waveform thereof. In FIG. 6, the rotation signal 10 from the speed generator 41 is input to the clock input of the counter 51, counted for a predetermined period by the enable signal 53 and the reset signal 54, and latched by the latch 52.

In FIG. 7, the counter 51 counts the rotation signal 10 while the enable signal 53 is high;
During the low period, the count result of each binary digit is latched signal 5.
5 to latch 52, and then reset by reset signal 54. As a result, the latch 52 receives the count result 11 for each enable signal period.
is output as a speed signal n.

FIG. 8 is a circuit diagram of the driver 7, which shows a pulse width modulated wave signal 1 modulated by the duty factor output from the microcomputer 6 shown in FIG.
2 is switched by the forward/reverse rotation signal 13 and output to the inverter 3.

FIG. 10 and FIG. 11 are diagrams showing an example of the starting characteristics of the motor speed control device according to the present invention, in which the dotted line shows the response before applying the present invention, and the solid line shows the response when the present invention is applied. There is.

Figure 10 shows a situation where the system was overbraking due to the excessive inertia factor of the load connected to the motor, resulting in a slow response as shown by the dotted line, starting from time 0.
Detect the rotation speed N _S1 up to t _s1 and find the acceleration α,
As a result of correcting the value of the integral system K _I , the time
The rise of the integral term from t _S1 increases, indicating improvement as shown by the solid line. The solid line descending to the right indicates the value of the proportional term, and when this value becomes zero, the system will settle down to a steady speed. Therefore, when comparing the time when the proportional term becomes zero, the settling time of the transient response is shortened from t ₂ to t ₁ according to the present invention.

Figure 11 shows a situation where the inertia factor of the load mentioned above was insufficient and the system was under braking, causing vibrations as shown by the dotted line.The rotation speed N _S1 was similarly detected and the integral coefficient
As a result of correcting the value of K _I , the rise of the integral term from time t _S1 becomes gradual, and the solid line shows that vibrations have been removed. Comparing the settling time of the transient response from the time when the proportional term becomes zero, it can be seen that the present invention similarly shortens it from t ₂ to t ₁ .

Note that the acceleration α may be determined from the time t _S1 during which the rotational speed N reaches a predetermined value from 0.

[Effects of the Invention] According to the present invention, the rotational acceleration α of the motor is detected and the integral constant of the PI control circuit is corrected, thereby quickly and accurately correcting fluctuations in the braking coefficient ζ of the motor speed control system. It is possible to prevent delays in motor response characteristics, hunting, etc. that occur due to changes in the inertia factor of the load acting on the motor, load torque, etc., and the motor can always be operated in an optimal braking state.

As a result, it is possible to increase the response speed of the system when starting up or changing the speed command, and at the same time prevent unnecessary vibrations such as hunting, and at the same time, it is possible to reduce the speed error to zero.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the basic configuration of the present invention, FIG. 2 is a block diagram showing FIG. 1 using a transfer function, FIG. 3 is a circuit diagram of one embodiment of the present invention, and FIG.
5 and 5 are flowcharts showing the calculations performed by the microcomputer according to the embodiment of the present invention, respectively.
Fig. 6 is a circuit diagram of the speed detection circuit used in the embodiment of the present invention, Fig. 7 is an operation waveform diagram of Fig. 6, Fig. 8 is a circuit diagram of the driver used in the embodiment of the present invention, and Fig. 9 8 is an operating waveform diagram, FIGS. 10 and 11 are diagrams showing a comparison of startup responses of the conventional device and the device of the present invention, and FIG. 12 is a block diagram of the conventional device. 1... PI circuit, 2... Motor, 3... Switching circuit, 4... Speed detector, 41... Speed generator, 42... Speed detection circuit, 5... Acceleration detector, 51... Counter, 52...Latsuchi, 6...
Microcomputer, 7...driver, N _C ...
...Speed command signal, 11...Speed signal n, 16...
Rotation direction command, 17...Start signal.

Claims (1)

[Claims] 1. A motor driven by a switching circuit, a speed generator that detects the rotational speed of the motor, a speed detection circuit that converts the output of the speed generator into a speed signal, and the speed detection circuit. means for generating a difference signal between the output of the output and the speed command value, and means for generating a proportional term proportional to the difference signal and a time integral term of the difference signal, wherein the proportional term and the time integral term are In the motor speed control method in which the switching circuit is driven by the sum of
or means for setting a predetermined time width when changing the speed command value, the rotational acceleration of the motor is determined from the change width of the speed signal within the time width, and the rotational acceleration is calculated based on the load torque or inertia of the motor. The coefficient of the time integral term is increased when the rotational acceleration is larger than a predetermined value predetermined by efficiency, and the coefficient of the time integral term is decreased when the rotational acceleration is smaller than the predetermined value. Motor speed control method. 2. A motor driven by a switching circuit, a speed generator that detects the rotational speed of this motor, a speed detection circuit that converts the output of the speed generator into a speed signal, and the output of the speed detection circuit and a speed command value. and means for generating a proportional term proportional to the difference signal and a time integral term of the difference signal; A speed control device for a motor configured to drive the motor, comprising means for setting a predetermined time width when starting the motor or changing the speed command value, and determining the speed of the motor according to the change width of the speed signal within the time width. Determine the rotational acceleration of the motor, and when the rotational acceleration is larger than a predetermined value predetermined based on the load torque and inertia factor of the motor, increase the coefficient of the time integral term, and when the rotational acceleration is smaller than the predetermined value. A speed control device for a motor, comprising: an arithmetic device that performs an arithmetic operation to reduce a coefficient of the time integral term.