JPH06351284A - Current controller for inverter - Google Patents

Abstract

PURPOSE:To allow sequential convergence of current error to 0 by delivering a switching vector function to a switching element depending on the current error, calculating a time function from the maximum absolute value of current error, and then delivering a zero vector switching function for correcting the current error at a timing corresponding to the value of the time function. CONSTITUTION:Detection signals ifu, ifv are fed from current detectors 3a, 3b to a CPU 9a along with a rotational position signal theta (n) from a position detecting circuit 7. The CPU 9a is externally fed with a torque command tauc(n). An I/O port 9h and a timer 8b constitute a correction means 18 having a current error correcting function. In other words, a time function Tz (n) is calculated from an absolute maximum value of current error and a corresponding value is set in a timer 18b. Every time when an interruption request signal INT2 is outputted from the timer 8b, a zero vector switching function is delivered to the main switching elements la-If of an inverter 10 in place of the switching vector function.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an inverter current control device for controlling the torque of an electric motor.

[0002]

2. Description of the Related Art An instantaneous value comparison current controller is known as a current controller for an inverter for controlling the torque of an electric motor. The example described in Japanese Patent Publication No. 3-50509 is an example. An outline of a conventional instantaneous value comparison type current control device for an inverter will be described below with reference to FIG. 1 showing an embodiment of the present invention.

In FIG. 1, an inverter 10 is a motor 12 having a three-phase structure with a DC power supply 1 m by switching operation.
The main switching elements 1a to 1f are supplied to the respective phase coils 14u, 14v, 14w. The main switching elements 1a to 1f are transistors. The currents supplied to the two-phase coils 14u, 14v are the current detectors 3a, 3
b, and the detected signals ifu and ifv are analog-to-digital converters 6a and 6a as current feedback signals.
It is input to the control calculation unit 9 via 6b. The encoder 4 outputs a detection signal each time the electric motor 12 rotates by a predetermined angle, and the position detection circuit 7 causes the electric motor 12 to output based on the detection signal.
The rotational position signal θ (n) is input to the control calculation unit 9. A torque command τc for instructing the rotation of the electric motor is given to the control calculation unit 9.
(N) is input from the outside. The control calculation unit 9 is the CPU 9
a. An interrupt signal INT1 is input to the CPU 9a from the timer 8a at regular intervals. In addition,
The (n) added after the code representing each signal means the value of the n-th sampling.

The control arithmetic unit 9 including the CPU 9a is shown in FIG.
It has a function as shown in 0. That is, in the current command generator 20, each time the timer 8a counts,
The current command ic (n) is calculated from the torque command τc (n) and the position signal θ (n), and the subtracter 22 calculates the current command i.
A current error Δi (n) between c (n) and the current feedback signal if (n) from the current detectors 3a and 3b is calculated, and the instantaneous comparison current control unit 21 calculates the current error Δi (n) according to the current error Δi (n). The switching vector function S (n) is output.

The switching vector function S (n)
Represents the switching states of the main switching elements 1a to 1f corresponding to the u, v, and w phase drive coils, respectively.
The case where the upper main switching element corresponding to each phase drive coil is on is represented by "1", and the case where the lower main switching element is on is represented by "0". For example, if the switching vector function S is “110” at a certain sampling point, and if the w phase exceeds the current error −Δi at the next sampling point, the switching vector function S is increased to increase the current of the w phase. As [111], the change of the pulse width modulation pattern by the inverter 10 is instructed. However, the switching vector function S is [111] or

[000] means that all the main switching elements on the upper side or all the main switching elements on the lower side are on, which means a zero vector, and the phase current is currently maintained and does not change. If the u phase exceeds the current error Δi at the next sampling point,
In order to reduce the u-phase current, the switching vector function S is set to [011] to instruct to change the pulse width modulation pattern.

Next, the operation of the above conventional example will be described with reference to FIGS.
Will also be described with reference to. The timer 8a manages the time of the current control software of the control calculation unit 9. That is, an interrupt request is issued to the CPU 9a by the interrupt request signal INT1 output from the timer 8a at a constant cycle T, and the interrupt processing shown in FIG. 11A is performed. C
The PU 9a samples the current feedback amount, torque command, and magnetic pole position to obtain τc (n), if (n), θ.
(N) is input, and the current command ic (n) is calculated from τc (n) and θ (n) by applying the equation (1).

[Equation 1]

Next, from ic (n) and if (n),
Is applied to calculate the current error Δi (n), and the switching function vector S (n) is obtained and output according to Δi (n).

[Equation 2]

Various switching elements 1a to 1f of the inverter 10 are turned on / off according to the switching function vector S (n), and the inverter 10 supplies voltages Vu, Vv,
Vw is output and the rotation of the motor 12 is controlled.

The calculation of the switching function vector S is shown in FIG.
The procedure is shown in 1 (b). First, the u-phase current error Δiu (n) is calculated from Equation 3, and it is determined whether Δiu (n) is larger than a predetermined threshold ΔI. If larger, the u-phase switching function vector Su (n) is calculated. [1] When Δiu (n) is smaller than the predetermined threshold value ΔI, the process directly proceeds to the next step. Next, it is determined whether or not Δiu (n) is smaller than a predetermined threshold value −ΔI, and if smaller, a u-phase switching function vector Su (n).
To

As [0], the following v-phase switching function vector S
Go to the step of calculating v (n), and if Δiu (n) is larger than the predetermined threshold value −ΔI, then the above Sv
Go to the step of calculating (n). The procedure for calculating Sv (n) and the w-phase switching function vector Sw
The procedure for calculating (n) is similarly performed.

[Equation 3]

By turning on / off each of the main switching elements 1a-1f of the inverter 10 according to the switching function vector S calculated as described above, the current error Δi is controlled to be zero. For example, the current iu
Is larger than the current command, the current error Δiu (n) becomes negative. Therefore, as described with reference to FIG. 11B, if Δiu (n) <− ΔI, 0 is selected as the switching function Su (n), and the switching element 1e is selected.
Is turned off, 1f is turned on, and Vu = 0 [V]. V
Assuming that v = Vw = V [V] is selected, u
The voltage Vun shown in Equation 4 is applied to both ends of the phase winding,
The current decreases.

[Equation 4]

FIG. 12 is a diagram showing the above operation. As shown in FIG. 12, when the nth sampling is performed by the interrupt signal INT1, the switching function vector S is [101], and as a result of calculation by the CPU 9a, Δiu (n) is larger than the threshold ΔI, and Δiv
If (n) is smaller than the threshold value -ΔI, the switching function vector S is set to [011] to decrease Δiu and increase Δiv. As a result of calculation at the next n + 1-th sampling, Δiu (n + 1), Δiv (n +
If both 1) and Δiw (n + 1) are within the threshold value, the switching function vector S is maintained at [011]. As a result of calculation at the next n + 2nd sampling, Δiw (n + 2) is larger than the threshold ΔI,
If u (n + 2) is smaller than the threshold value −ΔI, the switching function vector S is set to [110], and Δiw
Is decreased and Δiu is increased. In this way, each phase current is controlled so as to follow the target value.

[0012]

According to the above-described conventional current control device, the current error with respect to the target value of each phase current is sampled at a constant cycle, and the threshold is detected when the current error exceeds a predetermined threshold value. It is controlled so that it falls within the value. Therefore, if you set the target value of each phase current to a sine wave,
Each phase current can be controlled in a sinusoidal manner. However, according to the above-described conventional current control device, as a result of constantly increasing or decreasing each phase current and sampling and calculating each phase current at a constant period, the error with respect to the target value of a certain phase current exceeds the threshold value. In this case, since the phase current continues to increase or decrease until the error of the phase current exceeds the threshold value in the subsequent sampling, in principle each phase current does not converge to the target value, but to the target value. There is a problem that a large ripple is superimposed on each phase current as shown in FIG.

Moreover, when the inductance of the electric motor is small and the sampling time cannot be set short due to the processing of the CPU, the threshold value ΔI of the error current is set.
Has to be set to a large value, and there is a drawback that the ripple of each phase current becomes larger. As described above, when the ripple of each phase current is large, the noise generated by the electric motor becomes large, and the problem that high-precision speed control and position control cannot be performed is derived.

The present invention has been made in order to solve the above-mentioned conventional problems, and makes it possible to control each phase current so as to converge to a target value, and to prevent ripples superimposed on the phase current. An object of the present invention is to provide a current control device for an inverter that can be reduced in size, has low noise in the electric motor, and can perform speed control and position control with high accuracy.

[0015]

SUMMARY OF THE INVENTION The present invention comprises an inverter for driving and controlling an electric motor, a current detector for detecting an output current of the inverter, a position detector for detecting a rotational position of the electric motor, and a current detector. Current feedback signal,
The current command is calculated by inputting the torque command that commands the rotation of the motor and the position signal from the position detector for each count of the timer, and the current error is calculated from the current command and the current feedback signal. In a current control device for an inverter including a control operation unit for outputting a switching vector function to the switching element of the inverter, a time function is calculated from the maximum value of the absolute value of the current error, and zero is obtained at a time corresponding to the value. The control operation unit is provided with a correction means for outputting a vector switching function to the switching element instead of the switching vector function to correct the current error.

A time function may be calculated from the maximum absolute value of the current error and set in a second timer, and a zero vector switching function may be output for each count of this second timer.

[0017]

The control calculation unit calculates the current command and the current error at a constant cycle for each count of the timer, outputs the switching vector function according to the current error, and controls the switching elements of the inverter on / off. The control calculation unit also calculates a time function from the maximum value of the current error absolute value, and the correction unit of the control calculation unit sets the zero vector switching function to the switching element instead of the switching vector function at a time corresponding to the time function. To correct the current error. The zero vector switching function means maintaining the current state in which the current of each phase is neither increased nor decreased. The correction operation in which the correction means outputs the zero vector switching function is performed in the middle of the above-described operation in a constant cycle, and thereby the current error is controlled so as to converge to zero.

[0018]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of an inverter current control device according to the present invention will be described below with reference to the drawings. In FIG. 1, an inverter 10 has a DC power supply 1 m by switching operation and each phase coil 1 of a motor 12 having a three-phase configuration.
Main switching element 1 for supplying to 4u, 14v, 14w
a to 1f. Each switching element 1a-1f
The free wheeling diodes 1g to 1l are connected in parallel. The currents supplied to the two-phase coils 14u, 14v are detected by the current detectors 3a, 3b, and their detection signals ifu,
Ifv is input as a current feedback signal to the control calculation unit 9 via the analog / digital converters 6a and 6b and the input / output ports 9e and 9f.

The electric motor 12 has a rotor having a rotor magnet 16, and an encoder 4 constituting a position detector is connected to the rotor. Each time the rotor of the electric motor 12 rotates by a predetermined angle, the encoder 4 outputs a detection signal, and the position detection circuit 7 causes the electric motor 12 to output based on this detection signal.
Of the rotational position signal θ (n) is output to the control calculation unit 9 via the input / output port 9g. A torque command τc (n) that commands the rotation of the electric motor is also input to the control calculation unit 9 from the outside via the input / output port 9c. The control calculator 9 has a CPU 9a. An interrupt request signal INT1 is input to the CPU 9a from the timer 8a at regular intervals.

A second timer 8b is connected to the control operation section 9 via an input / output port 9h. The input / output port 9h and the second timer 8b constitute the correction means 18. The control calculation unit 9 including the CPU 9a has the function as described above with reference to FIG. 10, and also has the function of correcting the current error together with the correction unit 18. That is, from the maximum value of the current error absolute value to the time function T
z (n) is calculated, a time corresponding to the value is set in the second timer 8b, and the zero vector switching function is switched every time the second interrupt request signal INT2 is output from the second timer 8b. It outputs to the main switching elements 1a to 1f of the inverter 10 instead of the vector function.

The switching function S calculated by the control calculator 9 is output from the input / output port 9d. The switching function S includes a u-phase switching function Su (n), a v-phase switching function Sv (n), and a W-phase switching function Sw (n), and each switching function is a short-circuit prevention circuit 5a, 5b, 5c. Is input to the switching element corresponding to each phase to control ON / OFF of each switching element. The short-circuit prevention circuits 5a, 5b, 5c are composed of upper and lower switching elements corresponding to each phase, that is, 1a and 1b, 1
In order to prevent c and 1d, 1e and 1f from being turned on at the same time, they are turned on and off with a time difference to prevent destruction of the switching element.

The control / calculation unit 9 is configured by software so as to perform the operation described below. here,
It differs from the conventional example in that the second interrupt process is performed in addition to the first interrupt process. First, the timer 8a
Manages the time of the current control software of the control calculator 9. That is, an interrupt request is issued to the CPU 9a by the interrupt request signal INT1 output from the timer 8a at a constant cycle T, and the interrupt processing 1 shown in FIG. 2A is performed. The CPU 9a samples the current feedback amount, torque command, and magnetic pole position to obtain τc (n), if
(N) and θ (n) are input, and the current command ic (n) is calculated by applying Equation 1 from τc (n) and θ (n), and then ic (n) and if (n) Equation 2 is applied to calculate the current error Δi (n), the switching function vector S (n) is calculated according to Δi (n), and the absolute value of each phase current error Δi (n) is calculated. Is selected and the time function Tz (n) is calculated accordingly. The switching function vector S (n) thus calculated and the time function Tz (n) are output, and the interrupt processing 1 is ended.

The calculation procedure of the switching function vector S is shown in FIG. First, Equation 3 is applied to calculate Δiw (n), and then whether the positive or negative of Δiu (n) is determined, and when positive, the u-phase switching function vector Su (n) is [1],
When negative, Su (n)

Set to [0]. The v-phase and w-phase switching function vectors Sv (n) and Sw (n) are similarly set.

The calculation procedure of the time function Tz (n) is shown in FIG.
It shows in (c). Here, first, the absolute values of the three-phase current errors | Δiu (n) |, | Δiv (n) |, | Δiw (n)
| Is calculated, and the maximum value of these current error absolute values is selected and set as | Δi (n) | max. Equation 5 is applied from the maximum value | Δi (n) | max of the absolute value of the current error to calculate the time function Tz (n) according to the equation, and the processing ends. Since Δiu (n) + Δiv (n) + Δiw (n) = 0, when the sign of the current error of one phase is different from the sign of the current error of the other two phases, the absolute value of the current error is always other. Is larger than the current error of the two phases. Therefore, the absolute value of the current error whose sign of the current error is different from those of the other two phases is selected and
It may be i (n) | max.

[Equation 5]

According to the switching function vector S (n) output as described above, each of the switching elements 1a to 1f of the inverter 10 is turned on, as in the conventional example.
The inverter 10 is turned off and Vu, Vv, and Vw are output from the inverter 10. On the other hand, the time function Tz (n) is preset in the timer 8b, the timer 8b counts down from the preset value, underflows when the count value becomes 0, and the interrupt request signal IN
T2 occurs. Here, for the sake of convenience, it is assumed that when Tz (n) is preset, underflow occurs at Tz (n) [s]. CP by the interrupt request signal INT2
U9a performs the interrupt process 2 shown in FIG. That is, the zero vector switching function S ₀ =

[000]
Is output instead of the switching function vector S (n).

Zero vector switching function S ₀ = [0
[00] is output to output V from the inverter 10.
u = Vv = Vw = 0 [V] is output, and each phase current iu,
The changes in iv and iw are almost zero. The zero vector switching function S ₀ is to maintain the current state without increasing or decreasing each phase current, and the switching function [111] is also the zero vector switching function S ₀ . The above-described operation is repeated, and the current error Δi
Are controlled to focus on 0.

FIG. 4 is a diagrammatic representation of the above-described operation. As shown in FIG. 4, the interrupt process 1 is performed by the interrupt request signal INT1 output at a constant cycle, and at the same time as the interrupt process 1, the time function Tz (n) corresponding to the maximum absolute value of the current error signal is set. It is calculated and preset in the second counter 8b. Counter 8b
Counts down and underflows to output the second interrupt request signal INT2, thereby performing interrupt processing 2. The interrupt processing 2 is the zero vector switching function as described above.

[000] is output to maintain the current of each phase. By thus performing the interrupt process 2 between the interrupt process 1 and the next interrupt process 1, the current error is sequentially focused on 0 without penetrating the target value, and the ripple superimposed on the phase current is reduced. .

FIG. 5 shows each phase current in the above embodiment of the present invention. Comparing FIG. 5 with FIG. 13 showing each phase current of the conventional example, it can be seen that each phase current is finely controlled and the ripple superimposed on the phase current is obviously small. Therefore, the noise of the electric motor can be reduced,
In addition, highly accurate speed control and position control are possible. In the present invention, by outputting the zero vector switching function, the number of times the switching elements of the inverter 10 are switched is increased as compared with the related art. Therefore, in the present invention, the sampling time T is doubled and evaluated,
This is shown in FIGS.

Here, the derivation of the equation 5 will be described. In the model shown in FIG. 6, Vu = V [V], V
Consider a pattern of v = Vw = 0 [V]. u-phase current i
Focusing on u, an equivalent circuit as shown in FIG. 7 is obtained.
Here, R and L indicate the resistance and inductance measured between the terminal U and the neutral point N in FIG. 6, and E indicates the speed electromotive force. From FIG. 7, if the initial value of iu = 0, then Equation 6 is obtained.

[Equation 6]

Here, when the voltage drop due to the resistance and the speed electromotive force E are neglected, approximate expressions such as equations (7) and (8) are obtained.

[Equation 7]

[Equation 8]

If the error of the phase current iu (n) with respect to the current command ic (n) is Δiu (n), then Vu = V
After applying [V] and Vv = Vw = 0 [V], Δiu
The time tz during which 0 becomes 0 is as shown in Expression 9.

[Equation 9]

From the above results, the following can be seen. The switching vector function S (n) is the maximum value Δi of the current error.
There is a moment when Δimax = 0 when tz [s] elapses after selecting and outputting max (n) toward 0. This time tz is estimated, and after tz, V shown in FIG.
Zero vector switching function S ₀ such that x becomes ₀
Δimax is held at about 0 by outputting, and Δimax = 0 can be set at the next sampling time as shown in FIG. However, it is assumed that the time tz is shorter than the cycle T of the interrupt processing 1.

Therefore, if the preset value Tz (n) in the second timer 8b corresponds to the above-mentioned time tz [s] in a ratio of 1: 1, then Expression 5 is obtained from Expression 9.

[0034]

According to the present invention, in a current control device for an inverter, which outputs a switching vector function to a switching element of an inverter according to a current error and controls the switching element according to the switching vector function, Since the time function is calculated from the maximum value of the current error absolute value and the correction means for correcting the current error by outputting the zero vector switching function at the time corresponding to the value is provided, the current error is sequentially focused to 0, The ripple superimposed on the current is reduced. Therefore, the noise of the electric motor is reduced, and the accuracy of speed control and position control can be improved. Furthermore, there is an advantage that processing can be performed with relatively simple software and algorithms, and the processing time is short.

[Brief description of drawings]

FIG. 1 is a block diagram showing an embodiment of a current control device for an inverter according to the present invention.

FIG. 2 is a flowchart showing a part of the operation of the above embodiment.

FIG. 3 is a flowchart showing an operation of another part of the embodiment.

FIG. 4 is a timing chart showing the operation of the above embodiment.

FIG. 5 is a waveform diagram showing each phase current in the above embodiment.

FIG. 6 is a schematic diagram showing an example of an energized state of an electric motor controlled by the device of the present invention.

FIG. 7: Same as above, equivalent circuit in energized state.

FIG. 8 is an equivalent circuit that simplifies the energization state.

FIG. 9 is a graph showing the relationship between the maximum absolute value of the current error and the time until the current error becomes 0 in the embodiment of the present invention.

FIG. 10 is a functional block diagram of a control arithmetic unit in a conventional inverter current control device.

FIG. 11 is a flowchart showing the operation of a conventional inverter current control device.

FIG. 12 is a timing chart showing the operation of a conventional inverter current control device.

FIG. 13 is a waveform diagram showing each phase current in the conventional inverter current control device.

[Explanation of symbols]

 1a to 1f Main switching elements 3a, 3b Current detector 4 Encoder constituting a position detector 8a Timer 8b Second timer 9 Control operation unit 10 Inverter 12 Electric motor ifu, ifv Current feedback signal τc (n) Torque command θ (n ) Position signal S switching vector function

─────────────────────────────────────────────────── ───

[Procedure amendment]

[Submission date] December 3, 1993

[Procedure Amendment 1]

[Document name to be amended] Statement

[Name of item to be corrected] Claim 1

[Correction method] Change

[Correction content]

[Procedure Amendment 2]

[Document name to be amended] Statement

[Correction target item name] 0006

[Correction method] Change

[Correction content]

Next, the operation of the above conventional example will be described with reference to FIGS.
It will be described with reference to item 2 as well. The timer 8a manages the time of the current control software of the control calculation unit 9. That is, an interrupt request is issued to the CPU 9a by the interrupt request signal INT1 output from the timer 8a at a constant cycle T, and the interrupt processing shown in FIG. 11A is performed.
The CPU 9a samples the current feedback amount, the torque command, and the magnetic pole position to obtain τc (n), if (n), θ.
(N) is input, and the current command ic (n) is calculated from τc (n) and θ (n) by applying the equation (1).

[Equation 1]

[Procedure 3]

[Document name to be amended] Statement

[Name of item to be corrected] 0015

[Correction method] Change

[Correction content]

[0015]

The present invention SUMMARY OF], the inverter and the current detector and a current feedback signal from the current detector and current command Toka et current for detecting the output current of the inverter that drives and controls the electric motor In an inverter current control device comprising an error calculation and a control calculation unit that outputs a switching vector function to a switching element of the inverter according to the current error, a time function is calculated from the maximum value of the current error absolute value. Then, the control calculation unit is provided with a correction means for correcting the current error by outputting the zero vector switching function to the switching element instead of the switching vector function at a time corresponding to the value.

[Procedure amendment 4]

[Document name to be amended] Statement

[Correction target item name] 0017

[Correction method] Change

[Correction content]

[0017]

[Action] control arithmetic unit, current command, further calculates the current error, and outputs the switching vector function to on-off control the switching elements of the inverter in response to the current error. The control calculation unit also calculates a time function from the maximum value of the current error absolute value, and the correction unit of the control calculation unit sets the zero vector switching function to the switching element instead of the switching vector function at a time corresponding to the time function. To correct the current error. The zero vector switching function means maintaining the current state in which the current of each phase is neither increased nor decreased. The correction operation in which the correction means outputs the zero vector switching function is performed in the middle of the above-described operation in a constant cycle, and thereby the current error is controlled so as to converge to zero.

[Procedure Amendment 5]

[Document name to be amended] Statement

[Name of item to be corrected] 0029

[Correction method] Change

[Correction content]

Here, the derivation of the equation 5 will be described. In the model shown in FIG. 6, Vu = V [V],
Consider a pattern of Vv = Vw = 0 [V]. Focusing on the u-phase current iu, an equivalent circuit as shown in FIG. 7 is obtained. Here, R and L indicate the resistance and inductance measured between the terminal U and the neutral point N in FIG. 6, and E indicates the speed electromotive force. From FIG. 7, if the initial value of iu = 0, then Equation 6 is obtained.

[Equation 6]

Claims (2)

[Claims] 1. An inverter for driving and controlling an electric motor with an output current obtained by switching of a main switching element, a current detector for detecting an output current of the inverter, and a position detector for detecting a rotational position of the electric motor. , A current feedback signal from the current detector, a torque command for externally commanding the rotation of the electric motor, and a position signal from the position detector are input for each count of the timer to calculate a current command, and a current command A current controller for an inverter, comprising: a current calculation unit that calculates a current error from a current feedback signal and outputs a switching vector function to a switching element of the inverter according to the current error. Calculate the time function from the maximum value and set the zero vector switching function at the time corresponding to the value. Current control device of an inverter, characterized by comprising a correction means for correcting the current error is output to the switching element to the control arithmetic unit instead of the switch ring vector function. 2. The inverter according to claim 1, wherein a time function is calculated from the maximum value of the current error absolute value and is set in a second timer, and a zero vector switching function is output for each count of this second timer. Current control device.