WO1994027360A1

WO1994027360A1 - Method for controlling current for synchronous motor

Info

Publication number: WO1994027360A1
Application number: PCT/JP1994/000751
Authority: WO
Inventors: Hiroyuki Uchida; Yasusuke Iwashita; Takashi Okamoto; Hidetoshi Uematsu
Original assignee: Hiroyuki Uchida; Yasusuke Iwashita; Takashi Okamoto; Hidetoshi Uematsu
Priority date: 1993-05-18
Filing date: 1994-05-09
Publication date: 1994-11-24
Also published as: JPH06335279A

Abstract

An approximate function is obtained for indicating the relationship between the magnitude of driving current and the maximum current loop gain which does not cause undesired oscillation. A gain correction value k(I) is obtained from this function corresponding to the magnitude of the command current Ic. A PWM command obtained by the usual current loop processes (1) and (2) is multiplied by the gain correction value k(I) to execute a PWM process for the provision of a corrected PWM command, and thus the PWM signals are obtained to drive a motor. In this way, the band shortage of the current and velocity loops is eliminated while oscillation is being suppressed.

Description

Specification

Current control method for synchronous motor

Technical field

The present invention relates to a current control method for a synchronous motor, and more particularly, to a current control method for a synchronous motor that performs PWM (Pulse Width Modulation) control by software control (digital control).

Background technology

Figure 1 is a block diagram of the current loop of a synchronous motor that has been conventionally executed. In the figure, 1 is the integrator term in the IP control of the current loop processing, and K 1 is the integral gain. 2 is a term of the proportional gain K 2 in the current loop processing, 3 is a term for generating a PWM signal by a PWM command obtained in the current loop, and a transfer function is set to “1 j.” 4 is a mode. In the term representing the transfer function of the evening coil, R is the resistance of the coil, L is the inductance, and s is the Labrus operator. Only the block diagram of the loop is shown, and the other S phase, T phase, etc. are also the block diagram of the same current loop.

In the current loop processing shown in Fig. 1, the current command (torque command) I c output from the speed loop and the current command I re for the R phase obtained based on the mouth position and the feedback current I rf for the R phase The current deviation is calculated from this, the current deviation is integrated, and the value obtained by multiplying by the integral gain K 1 (the processing of item 1) is multiplied by the proportional gain K 2 by the R-phase feedback current I rf. (Processing of item 2) To find the R-phase PWM command. Then, a PWM signal is created by this PWM command (processing in section 3), and the inverter is driven by the PWM signal to control the R-phase current. The same applies to the processing of other S and T phases.

Optimum values of the integral gain K 1 and the proportional gain K 2 of these current loops are calculated from various parameters of the motor, that is, the values of the winding resistance, the inductance, etc. Is greatly influenced by the structure of the magnetic circuit, especially the structure of the magnetic circuit, and is determined by making appropriate corrections through evaluation tests with actual machines based on the value calculated as the optimum value. .

When modifying the values of the integral gain K 1 and the proportional gain K 2 in the above current loop, the values of the integral gain K 1 and the proportional gain K 2 must be adjusted within a range where the current loop is kept stable. It is desirable to increase it to the maximum in order to maximize the characteristics of this current loop. For this purpose, it is necessary to evaluate the stability of the current loop based on the maximum current. Generally, when a large current is applied, armature reaction occurs, magnetic saturation occurs at the teeth of the stator, and the inductance decreases. This is because there is considerable oscillation.

However, if a low gain setting is made to avoid oscillations when the maximum current is applied, the current loop itself will have insufficient bandwidth and the speed loop gain will not be raised. Becomes insufficient band, W / 7

-3 It becomes impossible to exhibit good servo characteristics, for example, the vibration during stoppage becomes large due to uneven feeding. This tendency becomes more pronounced as the magnetic flux density increases.

Disclosure of the invention

5 An object of the present invention is to provide a current control method for a synchronous motor that can solve the above-mentioned problems of the conventional example, suppress oscillation, and eliminate a shortage of band.

In order to achieve the above object, according to one embodiment of the present invention, there is provided a current control method for a synchronous motor that performs PWM control.

10 The current loop gain for the current command value is obtained by a function approximating the relationship between the current loop maximum gain that does not cause oscillation and the current magnitude, and the current loop processing is performed using the current loop gain to generate the PWM command. I'm asking.

In another aspect of the present invention, a synchronous

15 In the motor current control method, a correction value of the current loop gain with respect to the current command value is obtained by a function that approximates the relationship between the drive current and the maximum gain of the current loop in which oscillation does not occur. Current loop processing is performed by the current loop gain to obtain the PWM command.

Multiplied by the 20th correction value and output as a correction PWM command.

Further, in another aspect of the present invention, in a current control method for a synchronous motor that performs PWM control, a function similar to the relationship of the maximum gain of a current loop in which oscillation does not occur with respect to the magnitude of a drive current is used. The product of the current loop and the current command value The correction value of the minute gain and the correction value of the proportional gain are obtained, and in the processing in the current loop, the value obtained by multiplying the reference gain in which the proportional gain is set by the correction value of the proportional gain is obtained. In the integration processing, the value obtained by multiplying the reference integration gain by the correction value of the above integration gain is defined as an integration gain, and the value obtained by multiplying the integration gain by the current deviation is integrated to obtain an integration value. And ask for it.

Preferably, the correction value of the current loop gain of the current loop processing of the next speed loop processing cycle or the integration gain is determined by the current command value of the speed loop processing cycle obtained for each speed loop processing cycle. Determine the correction value of the gain and the correction value of the proportional gain.

More preferably, the actual current value is used in place of the current command value, and the actual current value is obtained by approximating the relationship between the magnitude of the drive current and the maximum gain of the current loop where no oscillation occurs with respect to the magnitude of the drive current. A current loop gain for the value is obtained, and a current loop process is performed by the current loop gain to obtain a PWM command. As described above, according to the present invention, the current according to the current command value or the actual current is determined by a function approximating the maximum gain at which oscillation does not occur with respect to the magnitude of the drive current. Since the loop gain is set and the current loop process is executed, no oscillation occurs, and the gain uses the maximum value for the current command value at that time, so that no band shortage occurs.

Brief explanation of drawings

Figure 1 shows a block diagram of a conventional current loop. FIG. 2 is a block diagram of a current loop according to the first embodiment of the present invention.

FIG. 3 is a block diagram of a current loop according to a second embodiment of the present invention,

Fig. 4 shows the maximum torque with respect to the motor drive current. Fig. 5 shows the winding inductance with respect to the motor drive current.

Figure 6 shows the maximum gain of the current loop where no oscillation occurs for the drive current.

FIG. 7 is a diagram showing an example of a gain function of a current loop with respect to a current approximating the curve of FIG. 6,

Figure 8 shows an example of the approximate function of the maximum current loop gain with respect to the current.

Fig. 9 is a flow chart of the processing in the current loop processing cycle when the processor of the digital servo circuit (software servo circuit) implements the first embodiment. The flow chart of the processing in the current lube processing cycle when the processor of the digital servo circuit (software servo circuit) implements the second embodiment. (1 O rpm) Rotation at (l O rpm)

FIG. 12 is a diagram of an experimental result in which feed unevenness is observed when the motor is rotated at a low speed (10 rpm) in the second embodiment of the present invention. Best mode for carrying out the invention

The relationship between the maximum torque with respect to the drive current and the relationship between the inductance with respect to the drive current and the curve of the magnetic saturation during one stage is shown by the curves. The general trends are generally shown in Figs. become. If the inductance for the current value shown in Fig. 5 is accurately obtained, the maximum value of the integral gain K1 and the proportional gain K2 of the current loop can be obtained by a predetermined calculation formula. In practice, the method of measuring the stability of the current loop by actually measuring the response to a step input and determining these gain values is generally adopted. ,

When such measurements are repeated at each current value and the maximum gain curve at which no oscillation occurs is obtained for each current value, the general tendency is as shown in Fig. 6. Therefore, a function is defined so that the current value becomes the gain indicated by this curve, and this function provides the maximum gain at which no oscillation occurs for the current value, and no oscillation occurs, and It is possible to obtain each gain of the current loop without causing band shortage.

However, as a practical matter, it is required that the setting method of the function be simple. Therefore, it is desirable to set a simple function that approximates the above curve. Thus, as an example of this function, in the present embodiment, the maximum gain curve for the current is represented by a simple polygonal line as shown in FIG. 7 (the polygonal line itself shown in FIG. 7 switches the gain as described later). (The value of the function k (I)) as the correction value for). And the book In the first embodiment of the present invention, as shown in the block diagram of FIG. 2, oscillation is generated by multiplying a PWM command obtained by ordinary current loop processing by a function value k (I) for gain correction. Try to get the maximum gain that doesn't change. FIG. 3 shows a second embodiment of the present invention.Integral gain K 1 and proportional gain K 2 of the current loop are given gain correction function values kl (I) and k2 (I), respectively. The maximum gain that does not generate vibration when multiplied is obtained.

That is, FIG. 2 in the first embodiment is different from FIG. 1 only in that a transfer function 5 of a gain correction function k (I) is added. Then, the corrected PWM command obtained by multiplying the PWM command output from the normal current loop processing by the value of the function k (I) for gain correction, which is a fraction of the current value, is obtained. It is designed to obtain a signal.

—On the other hand, the second embodiment shown in FIG. 3 is different from the conventional example of FIG. 1 in terms 1 ′ and 2 ′ corresponding to the terms 1 and 2. That is, in the second embodiment, the integral gain is changed to K1 · k1 (I), and the proportional gain is changed to K2 · k2 (I). In addition, in the integration process according to the term 1 ′, the gain K 1 · k 1 (1) (where kl (I) is the current deviation in the process cycle) is obtained by adding the current deviation £ between the current command and the feedback current in the process cycle. The gain is multiplied by a gain correction value (corresponding to the flow command Ic), and the integration process is performed. Therefore, the effect of the change in the gain correction function k (I) due to the change in the command current I c does not affect all integrated values, and B) Only the current deviation of the cycle processed by the CPU is affected. That is, there is only an effect corresponding to the current deviation in that cycle.

On the other hand, in the first embodiment shown in FIG. 2, a change in the value of the gain correction function k (I) affects the integrated value of the integrator. As a result, in the first embodiment of FIG. 2, there is a difference that the change of the PWM command is larger than the change of the current command as compared with the second embodiment shown in FIG.

In the present embodiment, the gain correction functions k U k 1 (1) and k 2 (I) are set as k (I) = k 1 (1) = k 2 (1), As shown in Fig. 7, the following M factor is applied to the flow command Ic (in Fig. 7, II and I2 are the current values of the gain switching ( II <1 2), which is the set value).

When the current command I c is O I c <I 1,

k (I) = 1

When the current command I c is I I i c <I 2, k (I) =

1 2- 1 1

… (1) When the current command I c is 1 2 ≤ I c <I iax,

k (I) =

Note that ct is 0 <α <1.

In this way, by setting the functions k (I), k1 (1), and k2 (1) for gain correction, the integral gain of the current loop, The proportional gain is obtained by multiplying the gain correction function shown in Fig. 7 by K1 times. K2 based on K1 and K2 set based on the small current. That is, when the command current I c is OI c and I 1, the integral gain and the proportional gain are Kl and K 2, respectively, and the command current I c is 12 ≤ I c <I max Then, the integral gain is K 1 · α and the proportional gain is Κ 2 · α. When I 1 ≤ I c <I 2, the value k (I) determined by the value of the command current value I c in the above equation is multiplied by the set values K 1 and K 2.

In addition, when the maximum gain curve shown in FIG. 6 is approximated by a polygonal line as shown in FIG. 7, the parameter required for setting the gain is determined by the integral at a small current serving as a reference. Only the gain K 1, the proportional gain K 2, the gain switching current values II and 12, and the parameter value α that reduces the gain are required.

For the gain switching current values II and 12, select and set the values according to the machine that uses the motor. In general, a servomotor used for feeding a machine tool requires smooth feed during machining, that is, at low current. On the other hand, when the current is large, it is mainly during rapid traverse acceleration / deceleration. Therefore, if the switching of the current loop gain occurs during machining, the machining accuracy may be affected. Therefore, the switching current values II and 12 should not be set too low. Fig. 8 shows an example where II is 50% of the maximum current and 12 is 75% of the maximum current. Next, the processing performed by the processor of the digital servo circuit (software servo circuit) in each current loop processing cycle will be described.

FIG. 9 is a flow chart of a process executed by the servo CPU (a processor of a digital servo circuit) when executing the first embodiment shown in the block diagram of FIG. It is one.

First, the current command (torque command) Ic obtained by the speed loop processing is read (step S), and the value of the gain correction value k is obtained from the magnitude of the current command Ic ( Step S2). That is, as described above,

When O I c <I l,

k = 1,

When I I ≤ I c <I 2,

1 — a

k =-I 1) +

1 2- 1 1

When I 2 ≤ I c <I max,

k = α

And

Next, the current commands I rc, I sc, and I tc of each phase are obtained, and the feedback currents I rf, I sf, and I tf of each phase are read (steps S 3 and S 4). Then, a current loop process is performed to obtain a PWM command (step S5). Then, the corrected PWM command is obtained by multiplying the gain correction value k obtained in step S2 by the obtained PWM command (step S6). A PWM command is output to the PWM circuit, and the processing in the cycle ends. Hereinafter, this process is repeated for each cycle.

FIG. 10 is a flowchart of a process for each current loop performed by the processor of the digital servo circuit in the second embodiment shown in FIG.

As in the first embodiment, the current command (torque command) Ic obtained by the speed loop processing is read (step T1), and the gain correction values kl and k are determined based on the magnitude of the current command Ic.磕 of 2 is obtained (step T 2). Note that the same function may be used as kl = k2, and the switching current values I 1 and I 2 and the value of the gain reduction parameter α may be changed as described above. Then, the values of these gain correction values kl and k2 are obtained from the value of the current command Ic.

Next, the current commands I rc, I s:, I tc of each phase are obtained, and the feedback currents I ri, I sf, It f of each phase are read (steps T3, 34), and the current loop processing is performed. And obtain the PWM command. In other words, the feedback current values I rf, I sf, and I tf are subtracted from the phase current command values I rc, I sc, and I tc, respectively, to find the current deviations £ Γ, £ 5, and ε 1 of each phase. Τ5), multiply the register A r, As, and At for each phase constituting the integrator by the gain correction value kl obtained in step T2 to the set integration gain K1. Further, a value obtained by multiplying each of the current deviations £ rεs and εt for each phase is added to obtain an integral value (step T6). Then, the set proportional gain Κ 2 is multiplied by the gain correction value k 2 obtained in step T 2, and the value obtained by multiplying each of the phase feedback currents I rf, I sf, and I tf is integrated by the integral value (Ar , As, At), respectively, to obtain a PWM command (step T7), output this PWM command, and end the processing of the cycle. Hereinafter, this processing is sequentially repeated.

In the above embodiment, the current command value Ic was used to determine the gain, but an actual current value flowing through the motor may be used instead of this current command value.

FIGS. 11 and 12 show examples of experimental results comparing feed irregularities of a feed shaft in a machine tool according to the method of the present invention and a conventional method. Fig. 11 shows the conventional method. Similarly, the proportional gain is also a value obtained by multiplying the calculated value by 06, and shows the unevenness of transmission when the motor is driven at a low speed of 10 rpm Indicates the time, and the vertical axis indicates the speed. If there is no transmission, the straight line shows the wavy speed due to the uneven transmission. Then, this transmission has a change of (SZIOOOCMrev or more).

'FIG. 12 illustrates a second embodiment of the present invention, in which the integral gain K 1 is 1.3 times the calculated value (the gain can be increased at low speeds because the gain is switched), and the proportional gain is increased. The motor K is driven at a low speed of l O rpm, with K 2 as 0.8 times the calculated value. The non-uniformity of the transmission was calculated, and as shown in Fig. 12, it can be seen that the non-uniformity of the transmission was suppressed to about (1/100) rev.

As described above, according to the present invention, the magnitude of the drive current (current command) is approximated by the function approximating the relation of the maximum current gain that does not cause oscillation. In other words, the current loop gain is determined according to the actual current value.Thus, while maintaining the stability of the current loop when the maximum current flows, the speed loop processing is performed by increasing the gain at low current. Since it is executed, the bandwidth of the current loop and the speed loop can be extended, and uneven transmission at low speed rotation and vibration at stop can be prevented.

Claims

The scope of the claims

In a current control method for a synchronous motor that performs PW Μ control, the relationship between the value of the drive current and the maximum value of the gain of the current loop that does not cause oscillation when the synchronous motor is driven with that value is determined in advance. Every

Electricity

The current gain when performing the lube process is changed to a value corresponding to the maximum value of the current loop gain obtained from the above relationship when the current command value or the feedback current value is replaced with the drive current.決定決定処理電流電流

W ΜA method for controlling the current of a synchronous motor, in which a command is obtained.

In the current control method of a synchronous motor that performs PW Μ control, a function approximating the relationship between the magnitude of the drive current and the maximum gain of the current loop where oscillation does not occur is used as the current loop gain for the current command value. A current control method for a synchronous motor, characterized in that a current loop process is performed by the current loop gain to obtain a PWM command. The synchronous circuit according to claim 1 or 2, wherein a current loop gain of current lube processing during the next speed loop processing cycle is determined based on a current command value obtained for each speed loop processing cycle. Electric motor current control method.

In the current control method of a synchronous motor that performs PW Μ control, a function approximating the relationship between the magnitude of the drive current and the maximum gain of the current loop where oscillation does not occur is used as the current loop gain for the current command value. Find the correction value of A current control method for a synchronous motor, comprising: obtaining a PWM command by performing a current loop process by using a current loop gain of the present invention, multiplying the PWM command by the above correction value, and outputting as a corrected PWM command.

5. The current loop gain correction value for the above current command value is determined based on the current command value of the speed loop processing cycle obtained in each speed loop processing cycle, and the current loop processing of the next speed loop processing cycle. The current control method for a synchronous motor according to claim 4, wherein the correction value of the current lube gain is determined.

6. In the current control method of the synchronous motor that performs PWM control, the current with respect to the current command value is obtained by approximating the relationship between the drive current and the maximum gain of the current loop in which oscillation does not occur with the magnitude of the drive current. The correction value of the integral gain of the loop and the correction value of the proportional gain are obtained, and in the processing in the current loop, the value obtained by multiplying the reference gain for which the proportional gain is set by the correction value of the proportional gain is used. In the integration process, the value obtained by multiplying the reference integration gain by the correction value of the integration gain is set as an integration gain, and the value obtained by multiplying the integration gain by the current deviation is integrated. A current control method for a synchronous motor, characterized in that it is obtained as an integral value.

7. The correction value of the integral gain of the current loop and the correction value of the proportional gain with respect to the above current command value are determined based on the current command value of the speed loop processing cycle obtained for each speed loop processing cycle. Of the speed loop processing cycle 7. The current control method for a synchronous motor according to claim 6, wherein the correction value of the integral gain and the correction value of the proportional gain in the current loop processing are determined.

8. The function approximating the relationship of the maximum gain of the current loop where oscillation does not occur to the magnitude of the drive current is as follows: the drive current is divided into several regions, and each divided region has a linear equation. The current control method for a synchronous motor according to any one of claims 2, 4, and 6, comprising:

9. The current control method for a synchronous motor according to claim 6, wherein the correction value of the proportional gain and the correction value of the integral gain are determined equally.

10. Instead of calculating the current loop gain for the above-mentioned current flow command value and performing the current loop processing by the obtained current loop gain, the current loop gain for the actual current value is calculated and the current loop gain is calculated by the obtained current loop gain. 7. The current control method for a synchronous motor according to any one of claims 2, 4, and 6, wherein a loop process is performed.