JPH037086A - Motor controlling method - Google Patents

Abstract

PURPOSE:To perform a high-accuracy positioning by narrowing the width of acceleration and deceleration control regions provided on both sides of a target control line passing the origin of rectangular coordinates formed by a motor speed omega and the deviation theta of the target and present positions. CONSTITUTION:The difference theta between position data theta of a pulse generator 9 and target position data thetas is obtained by a counter 10 and inputted to a digital signal processing circuit DSP12. In the DSP12, the output of said counter 10 is saturated 13 and a rotational speed omega is obtained by a speed detection circuit 14 and inputted to an arithmetic circuit 15 for the purpose of obtaining S by an expression shown in said circuit 15. Results operated by a triangular wave generation circuit 16 and comparison circuits 17, 18 is outputted from the DSP12 and inputted to ROM 5 in co-operation with clock 21, timer 19 and counter 6. The ROM 5 controls an inverter 2 by its built-in program. Thus, acceleration and deceleration regions generated on both sides of a target control line passing the origin of rectangular coordinates formed by said speed omega and the deviation theta is narrowed and positioned with high accuracy.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a motor control method that can accurately determine the rotational angular position of a motor or the position of a moving object connected to the motor.

[Prior Art] In order to stop a motor or a moving object connected to the motor at a predetermined position, the difference between the current position and the predetermined position is determined, and if this difference is large, the speed of the motor is increased to reduce the difference. It has already been done to reduce the speed of the motor when is small.

It also reduces the speed proportionally to the difference (deviation) in position #
In order to stably achieve I control, a curve showing the ideal relationship between deviation and speed should be determined in advance, and the position and speed should be converged to zero along this ideal curve. It is disclosed in the conference proceedings Nα1516 (page 2032).

In addition, the ROM is used to control the PWM inverter for the AC motor.
This method is disclosed in Japanese Unexamined Patent Publication No. 62-207196.

[Problems to be Solved by the Invention] However, there is a need for a motor control system that is more stable and accurate.

Therefore, an object of the present invention is to provide a motor control method that can meet the above requirements.

[Means for Solving the Problems] In order to achieve the above object, the present invention includes the following steps:
The positional deviation Δθ consisting of the difference between the current position and the target position is the first
A target control line (sliding curve) passing through the zero point is set in coordinates with the motor speed ω being the axis of the motor and a second axis perpendicular to the first axis, position data indicating the current position is obtained, and The position data and the target position are compared to determine the position difference Δθ, speed data indicating the speed ω of the motor is obtained, and the coordinate position determined by the position deviation (Δθ) and the detected speed ω is In the motor control method, the motor is controlled by i'l/l so as to obtain a positional error with respect to a target control line and correct the difference while converging the positional deviation to zero along the target control line. An additional 31 vJm area is provided on one side of the line, a deceleration control area is provided on the other side of the proportional control area, and a proportional control area is provided on both sides of the target control line, and the width is set as above. The present invention relates to a motor control method characterized in that the positional deviation is narrowed as the positional deviation becomes smaller. Note that it is desirable to add the integral value of this or a signal corresponding to this to the difference signal at the point where the positional deviation becomes zero.

[Function j] By widening the proportional control region in the region where the positional deviation is large and narrowing it in the region where the positional deviation is small, stable speed control is possible in the region where the positional deviation is large, and the gain is reduced in the region where the positional deviation is small. This is equivalent to increasing , which improves responsiveness and enables highly accurate positioning.

Furthermore, by adding the integral values, it becomes possible to obtain the necessary torque in a region where the deviation is near zero.

[Embodiment] Next, a speed control method for a three-phase motor according to an embodiment of the present invention will be described. In FIG. 1, a motor 1 consisting of a three-phase induction motor is equipped with a three-phase inverter 2 capable of PWM control.
is connected. As shown in FIG. 2, the inverter 2 has a DC power supply 3 connected to a switching element At
, A2, Bl, B2, CI, and C2 are bridge-connected.

The six switch elements A1 to C2 are turned on and off in response to control signals supplied from the drive circuit 4. In addition,
Three switch elements on the upper side of inverter 2, l, B1
, CI and the lower three switch elements A2, B2, C
2 operate inversely to each other, so if one control is specified, the control of the entire inverter is specified. here,
The inverter control state is specified by the first, second, and third signals A, B, and C read from the ROM (read-only memory) 5, and the signals A, B, and C are at high level, that is, logic "1". When the switch elements A1, B1, and C1 are on, when they are at a low level, that is, logic "O", the switch elements A1, B
1. CI is off.

ROM5 is PWM for controlling inverter 2 by PWM.
A switching pattern (unit vector data) is written in advance. This ROM5 has a forward rotation PWM pattern memory M1, a forward rotation zero vector memory M2, a reverse rotation PWM pattern memory M3, and a reverse rotation zero vector memory M4. Each memory M1 to M4 is O to 511
Each address has 512 addresses up to 1, and is addressed by the value of the 9-bit binary column 6a of the up/down counter 6. However, one of the four memories M1 to M4 is selected, and only the output of this selected memory is effectively used for controlling the inverter 2. In order to perform this selection, the ROM 5 has a zero vector selection control signal input terminal 7 and a forward/reverse rotation selection control signal input terminal 8.

When the zero vector selection control signal input terminal 7 is at logic "1", one of the memories M1 and M3 is selected, and when it is at the logic "0", one of the memories M2 and M4 is selected. Further, the forward/reverse rotation selection control signal input terminal 8 is set to “1”.
”, one of the memories M1 and M2 is selected, and in the case “O”, one of the memories M3 and M4 is selected.

Now, if the 9 bits on line 6a are represented by AO to A8, the input bits to input terminal 7 are represented by A9, and the input bits to input terminal 8 are represented by AIO, then the address is specified by the 9 bits from AO to A8. Ru. Mana A9, A10 [11
], the first memory M1 (forward PWM switching pattern) is selected, and when [01], the second memory M2 is selected.
(zero vector for forward rotation) is selected, the third memory M3 (reverse PWM switching pattern) is selected when it is [10], and the fourth memory M4 is selected when it is [00].
(zero vector for reversal) is selected.

A position detector 9 consisting of a pulse generator is coupled to the motor 1, and this output line is connected to a 24-bit up/down counter 10 in order to increase the output voltage of the inverter 2 by 7111 m and the output voltage of the inverter 2 to $18. The connected zero position detector 9 consists of an encoder that generates 1,296,000 pulses per revolution, and generates position data θ.

The counter 10 functions as a comparison circuit and generates a deviation Δθ between the position command data θS given from the line 11 and the detected position data θ.

The 24-bit output line of counter 10 is connected to a digital signal processing circuit (DSP) 12. The digital signal processing circuit 12 consists of a microcomputer,
It is a conflict that operates according to predetermined software. Note that in order to facilitate understanding of the operation of the digital signal processing circuit 12, the digital signal processing circuit 12 is constructed using the saturation circuit 1.
3, a speed detection circuit 14, an arithmetic circuit 15, a triangular wave generation circuit 16, and first and second comparison circuits 17 and 18.
It is shown separately from the absolute value circuit 18a.

The saturation circuit 13 limits the deviation Δθ input to the arithmetic circuit 15, and converts the 24-bit output of the counter 10 to 1.
A deviation Δθ limited to 6 bits is given to the arithmetic circuit.

The speed detection circuit 14 consists of a one-sample delay circuit 14a and a subtraction circuit 14b, and obtains speed data ω based on a change in position per unit time.

Operation port F#115 is 5=ci ・Δθ when Δθ>0
-C2・ω2 is calculated, and when Δθ<0, 5=CI
・Calculate Δθ102 ・ω2, however, CI,
C2 is a constant and ω is the rotation speed.

This calculation output S indicates whether or not the ideal rotational speed ω is obtained at an arbitrary positional deviation Δθ, that is, the output S includes information indicating the difference between the ideal speed and the actual speed in positioning control. There is. Figure 3 shows the relationship between the position error Δθ on the X axis and the speed ω on the X axis. The target control line shows the ideal value when Δθ>0, and 5=CI ・Δθ −C2・ω2=0 The line E shows in principle the change in the actual positional deviation Δθ and the change in the speed ω.6 In this embodiment, control is performed so that the actual line E is along the target control line. do. Therefore, the arithmetic circuit 15 in FIG. Data corresponding to the difference between the 3'II line and the actual dE is output.

The control method shown in FIG. 3 is known as a sliding mode method.

The triangular wave generation path 16 includes an amplitude modulation circuit 16a, and generates digital triangular waves of different amplitudes according to changes in positional deviation. Note that the digital triangular wave has a repetition frequency of 2 kHz. FIG. 4 shows the relationship between the amplitude of the triangular wave and the positional deviation Δθ. As is clear from this, as the positional deviation Δθ becomes smaller, the amplitude of the triangular wave also becomes smaller.

The first comparison circuit 17 determines whether the positional deviation Δθ is positive or negative;
When it is positive, it outputs "1" indicating forward rotation, and when it is negative, it outputs "0" indicating reverse rotation. This forward/reverse rotation (F/B) signal is applied to terminal 8 of the ROM.

The second comparison circuit 18 compares the triangular wave with the absolute value of the calculation output S, and outputs 10-bit data indicating the magnitude relationship between the triangular wave and the output S. That is, as shown in FIG. 9(A), the triangular wave Vc is compared with the absolute value of S obtained from the absolute position 8a, and digital data corresponding to FIG. 9(B) is output.

From one timer that operates in the same manner as the monostable multi-by-break connected to the output stage of the comparator circuit 18, the
) Binary data consisting of "1" and "0" is obtained. The output of the timer 19 of "1" indicates a period for operation, and "0" indicates a period of stop (zero vector generation). The output line of the timer 19 is connected to the zero vector selection control signal input terminal 7 of the ROM 5. AND gate 2
connected to o.

The clock generator 21 generates a 30kHz clock pulse by A
A 2 MHz clock pulse is applied to the clock input terminal of the counter 6 via the ND gate 20 and to the D/A converter 19. A 30 kHz clock pulse is applied to the counter 6 through the AND gate 20 when the run/stop signal is '1', and the address in the ROM 5 is incremented.

[Contents of ROM] Data is written in the ROM 5 as shown in principle in FIG. That is, each memory M1 to M4 of the ROM 5 has addresses 0 to 511, and the normal PWM pattern memory M
For example, the voltage vector v6, ■2 is in addresses 0 to 3 of 1.
, V6, and V2 are written in order, and the zero vector V7 is written to addresses 0 to 3 of the zero vector memory M2 for normal rotation.
, VO1V7, and VO are written in order, data of voltage vectors V1, v5, ■1, and ■5 are written in order in addresses 0 to 3 of the reverse PWM pattern memory M3, and zero is written in the reverse zero vector memory M4. Vector V
Data of O1v7 and VO1v7 are written in order. Vector data is also written to the remaining addresses 4 to 511 using the same principle as addresses O to 3. Since the vector data of each address in FIG. 6 shows the principle, it differs from actual data. Now, addresses 0 to 84 (O'''' to 60''' to 6 of the forward PWM pattern memory M1)
The voltage vector data at the 0' interval is v6, v6
, VO, VO, Vl, v2, Vl, v2, Vl, v2,
v6, VO, VO, VO, Vl, v2, Vl, Vl, V
l, v2, VO, VO, VO, VO, v2, Vl, Vl
, Vl, Vl, Vl v2, Vl, Vl, Vl, Vl,
Vl, v2, Vl, v2, Vl, v2, v2, Vl, V
l, ■2, Vl, Vl, Vl, Vl, Vl
, Vl , Vl, Vl, Vl, v2, Vl, v2, Vl
, Vl, Vl, Vl, ■3, V3, V3, V3, Vl,
Vl, Vl, v2, Vl, Vl, V3, V3, V3, V
3, Vl, Vl, Vl, Vl, Vl, Vl
, V3, V3, V3, V3 [voltage vector] FIG. 7 shows six voltage vectors v1 to V6 and two zero vectors VO, Vl, switch elements AI + 81, CI of inverter 2. Possible switching states are <000), (001), <010), (011)
, (Zoo), (101) (110), (111) 8
Therefore, we can define this as VO, Vl, v2, v3, v4,
In the device of this embodiment, the voltage vector VO-V7 is written in the ROM 5, and this is output as #J control data (A, B, C). By combining the eight vectors vO to V7, a sinusoidal output voltage and rotating magnetic field vector can be obtained.

[Vector Selection] Fig. 8 shows the selection of the voltage vector to obtain the rotating magnetic field vector φ1.
In the 30' interval, the sixth and second vectors V6, Vl,
The second and third vectors V2 and V in the 30” to 90” section
3. Third and first vector V in the 90° to 150° interval
3, Vl, 1st and 5th vector v1 in 150° to 2100° interval, V5. 5th and 4th vector V5 in 210° to 270° interval, V4. 6 vectors V4 and VO are selected.

In principle, in the 330° to 30° section of FIG. 8, 6 and 2 are selected as significant vectors, and the zero vector V7 is selected when vector rotation is stopped. Motor 1
When rotating in the normal direction, the rotating magnetic field vector φ1 is rotated in the direction indicated by UP in FIG. 8, and when rotating in the reverse direction or braking, the rotating magnetic field vector φ1
It is rotated in the direction indicated by OWN.

[Operation] The position data θ obtained from the position detector 9 and the target position data θS are compared by the counter 10, and a position deviation of 0 is obtained, and this value is significantly different from 0 as shown in Fig. 3. If so, the motor 1 must be rotated at high speed to bring it closer to the target position. In order to quickly and stably achieve this, the position deviation data Δθ and the speed data ω obtained from the speed detection circuit 14 are used in an arithmetic circuit. 15, and the value of S is determined according to the formula shown in the arithmetic circuit 15. As shown in Figure 3, if Δθ is significantly far from 0 at the starting point, the value of S will of course not become zero, and the value of S will cross the triangular wave shown in Figure 9 (A). It is located above this without. Therefore, the timer 19 continuously sends out "1", and the motor 1 is continuously driven with substantially no stop period (zero vector period). That is, by addressing the counter 6, the contents of the first memory M1 of the ROM 5 are read out one item at a time. This causes the speed of motor 1 to change to the third speed.
It changes as shown by line E in the figure. That is, as the angular position of the motor 1 approaches the target position, the rotational speed increases, and finally crosses the target control line.

Note that when the line E comes close to the line, the difference between the two becomes small, and the calculation output S comes to cross the triangular wave Vc shown in FIG. 9(A). As a result, the duty ratio between the operating period and the stopping period changes according to the change in the calculation output S.
A proportional control state is obtained. The speed of motor 1 jumps to the left and right of the target control line in Figure 3 due to inertia, however,
In this embodiment, the amplitude of the triangular wave Vc is set large as shown in FIG. ), and a stable control state can be obtained.

In FIG. 3, the area between the two dotted lines is the proportional area. As is clear from FIG. 3, the speed ω of the motor l is controlled along the target control line. As the motor 1 continues to rotate, the positional deviation Δθ gradually becomes smaller. At the same time, the amplitude of the triangular wave Vc gradually decreases as shown in FIG. 9(A). Since there is little possibility that the control will become unstable in a low speed region, the amplitude of the triangular wave Vc is reduced and the gain is equivalently increased to improve responsiveness and accuracy.

Although the digital signal processing circuit 12 is shown in FIG. 1 by analog analogy, control is actually executed by software shown in FIG.

As is clear from this flowchart, in the region where Δθ is smaller than the near-zero value θ0, S2 = 2(Δθ)fΔ
The integral value of θdt is determined. That is, the integral value of the positional deviation Δθ is determined. Instead of this integral, the integral value of S is 82
, and in the region of Δθ and θ0, U=S+
32 is used to determine whether the rotation is forward or reverse, and the comparison circuit 18 compares the absolute value of the current U with the triangular wave. S is the integral value S2
If you add , the apparent difference in clarity becomes larger, and the difference between the triangular wave and the U
The period in which the comparison output is "1" becomes longer, and the driving period of the motor 1 becomes longer. This increases the torque of the motor 1 and prevents it from stopping unnecessarily in the low speed range.

If the comparison output of the first comparison circuit 17 is at a high level "1" as shown before t5 in FIG. 9(C), this is input to the counter 6, and the counter 6 increases during this period Operate. Timer 1 of the output stage of the second comparison circuit 18
From now on, when the triangular wave VC is higher than S (t1 to t2)
Generates a low level output "0" at the opposite time (t2 to t3)
A high level output “1” is generated. During the period from t1 to t2, [A9 AIO] = [0
1], the normal rotation zero vector memory M2 is selected, and as in t2 to t3, [A9 AIO] = [11
], the normal rotation PWM pattern M2 is selected.

Note that from t1 to t2, the clock pulse does not pass through the AND gate 20 and the counter 6 is not incremented, so the same address continues to be specified.

On the other hand, from t2 to t3, the clock pulse is AND gate 2
It passes through 0 and becomes the input signal of the counter 6. As a result, the value of 9 bits AO to A8 of the counter 6 is increased by the up operation, and the addresses of the memory M1 are sequentially designated.

However, when the waveform of FIG. 9(B) becomes low level at time t3, the clock input to the counter 6 is inhibited, and the counter 6 retains the address designation at this time.
As shown in the figure, when memory M2 is selected while vector V6 of memory M1 is being read out at address 2, zero vector V7 (111
) is selected. The zero vector V7 continues to occur while the waveform of FIG. 9(B) is at a low level. When the tarok pulse is input to the counter 6 again and the output of the counter 6 is incremented by one step, the zero vector Vl (010) at address 3 of the normal PWM pattern memory M1 is selected. There are two types of zero vectors, VO (000) and V7 (111), and the vector that requires less switching of the switching elements A1 to C2 is selected. counter 6 is 10
When the binary numbers corresponding to the base numbers 0 to 511 are generated, all voltage vector data from 0 to 360° of the normal rotation PWM pattern is read out, three-phase approximate sine wave voltage is generated from the inverter 2, and A rotating magnetic field vector that is close to a circular locus is generated in the motor l.

In such control, as the difference between the target position and the detected position becomes smaller, the low level period of the 91st J(B) waveform becomes relatively longer, and the period during which the zero vector is selected becomes longer.

Note that if t2 in FIG. 9 is 0 degrees, vector v6, as shown in FIG.
VO, v6, ■6, ■2, v2 occur at a constant cycle with a constant width TP, and then the zero vector vO occurs from t3 to 74.
Occurs between M.

According to the present mJ method, ultra-high precision positioning of 1 second can be achieved with d reduction of force.

[Modifications] The present invention is not limited to the above-described embodiments, and, for example, the following modifications are possible.

(1) The digital signal processing channel 12 shown in FIG. 1 can be constructed with an analog circuit.

(2) The speed data ω may be obtained using a speed detector.

(3) As shown in FIG. 11, the ideal gap between Δθ and ω (target control line) may be made into a straight line.

(4) Instead of obtaining speed data using the same method during the entire period, in the area around the target position, speed data is obtained using the method shown in 1 [J], and the position deviation is small or disappears near the target position. In the low-speed region, speed detection may be performed by a method of measuring the number of clocks during the position deviation change period to improve accuracy. This can improve convergence with respect to the sliding curve near the target position. That is, in the # speed region, the period of the pulse obtained from the position detector 9 may be measured by counting clock pulses, and speed data may be obtained based on this and provided to the digital processing circuit 12.

[Effects of the Invention] According to the present invention, highly accurate positioning can be achieved.

[Brief explanation of the drawing]

Fig. 1 is a block diagram showing a motor control method according to an embodiment of the present invention, Fig. 2 is a circuit diagram showing details of the inverter shown in Fig. 1, and Fig. 3 is a position difference Δθ and speed in the method shown in Fig. 1. Figure 4 is a diagram showing the relationship between ω and ω using coordinates, Figure 4 is a diagram showing the relationship between the amplitude of the triangular wave and the position difference Δθ in the method shown in Figure 1, and Figure 5 is a flowchart showing the software for the method shown in Figure 1. , Fig. 6 is a diagram showing a part of the contents of the ROM in Fig. 1 in principle, Fig. 7 is a diagram showing voltage vectors, Fig. 8 is a diagram showing rotating magnetic field vectors, and Fig. 9 is a diagram similar to Fig. 1. Figure 10 is a diagram showing the state of each part of
FIG. 11, which is a diagram showing a part of FIG. 1...Motor, 2...Inverter, 5...ROM
, 9...Position detector, 10...Counter, 12...
・Digital signal processing concave path, 16...triangular wave generation circuit.

Claims (1)

[Claims] [1] In order to set the rotational angular position of the motor or a position corresponding thereto to the target position, a positional deviation (Δθ) consisting of the difference between the current position and the target position is set as a first axis, Motor speed (
Set a target control line (sliding curve) passing through the zero point in coordinates with ω) as a second axis orthogonal to the first axis, obtain position data indicating the current position, and combine the position data and the target position. The position deviation (Δθ
), obtain speed data indicating the speed (ω) of the motor, and calculate the difference in position between the coordinate position determined by the position deviation (Δθ) and the detected speed (ω) and the target control line. and controlling the motor so as to converge the positional deviation to zero along the target control line while correcting the difference, the method comprising: determining an acceleration control area on one side of the target control line; and a deceleration control area on the other side, and proportional control areas on both sides of the target control line, and the width of the proportional control area is narrowed as the positional deviation becomes smaller. Control method. [2] In order to set the rotation angle position of the motor or a position corresponding thereto to the target position, the position deviation (Δθ) consisting of the difference between the current position and the target position is set as the first axis, and the motor speed (
Set a target control line (sliding curve) passing through the zero point in coordinates with ω) as a second axis orthogonal to the first axis, obtain position data indicating the current position, and combine the position data and the target position. The position deviation (Δθ
), obtain speed data indicating the speed (ω) of the motor, and calculate the difference in position between the coordinate position determined by the position deviation (Δθ) and the detected speed (ω) and the target control line. In the motor control method of controlling the motor so as to converge the positional deviation to zero along the target control line while correcting the difference, A motor control method characterized by determining an integral value and stopping the motor at a target position while controlling the motor in proportion to the sum of the integral value and the signal indicating the difference. [3] In order to set the rotation angle position of the motor or a position corresponding thereto to the target position, the position deviation (Δθ) consisting of the difference between the current position and the target position is set as the first axis, and the motor speed (
Set a target control line (sliding curve) passing through the zero point in coordinates with ω) as a second axis orthogonal to the first axis, obtain position data indicating the current position, and combine the position data and the target position. The position deviation (Δθ
), obtain speed data indicating the speed (ω) of the motor, and calculate the difference in position between the coordinate position determined by the position deviation (Δθ) and the detected speed (ω) and the target control line. In the motor control method, the motor is controlled to converge the positional deviation to zero along the target control line while correcting the difference, at least in a low speed region of the motor, for speed detection or position detection. A motor control method characterized in that the speed data is obtained by measuring the period of a signal obtained in relation to the motor as a clock pulse. [4] The motor control method according to claim 1, wherein the motor is an induction motor. [5] The motor control method according to claim 4, wherein the speed control of the motor is performed by a pulse width modulation inverter.