The present invention relates to a motor driving apparatus that drives an induction motor at variable speeds, and relates in particular to a driving apparatus that can output a high torque at the time a motor is activated.
Recently, variable-speed driving of motors using an inverter has been developed to save energy and prevent global warming, or to provide improved efficiency for production lines.
That is, the speed controller 11, the flux instruction calculator 12 and the current instruction calculator 13 constitute current instruction calculation means for calculating a current instruction value based on a deviation between the speed instruction value ωr* and the estimated speed value ωr^. Since the control process is performed by dividing current elements into those along the axis (d axis) parallel to magnetic flux elements and those along the perpendicular axis (q axis), this process is also called a vector control process. A Q-axial current controller 14 and a d-axial current controller 15 calculate a q-axial voltage instruction Vq* and a d-axial voltage instruction Vd*, so that a torque current detection value IqFB=the torque current instruction Iq* and an excited current detection value IdFB=the excited current instruction Id* are established. These controllers 14 and 15 constitute current control means for controlling a current based on a current instruction value. The d-axial voltage instruction Vd* and the q-axial voltage instruction Vq* are converted into three-phase AC voltage instructions Vu*, Vv* and Vw* by employing a phase θ1, which is obtained by performing the integration of a frequency instruction value ω1* for an inverter that will be described later. Then, a PWM inverter 2, which is connected to a three-phase AC power source 1, performs PWM modulation for the three-phase instructions, and transmits the results, as the output voltage of the three-phase AC inverter, to an induction motor 3. The PWM inverter 2 performs switching by employing a semiconductor device, such as an IGBT. An induction voltage calculator 19 employs, for example, the following expressions (1) and (2) to convert the d-axial voltage instruction Vd* and the q-axial voltage instruction Vq* into motor induction voltages Ed and Eq. In these expressions, r1 denotes the primary resistance of a motor, Lσ denotes the sum of primary reduced values of leakage inductances of the motor, and P denotes a differential operator (d/dt).
Ed=Vd*−r1×Id−Lσ×P×Id+ω1×Lσ×Iq (1)
Ed=Vq*−r1×Iq−Lσ×P×Iq−ω1×Lσ×Id (2)
ω1*=Eq/Φ* (3)
The induction voltage calculator 19 and the frequency instruction calculator 20 constitute frequency instruction calculation means that employs the output voltage instruction value to calculate the frequency instruction value ω1* in the above described manner. It should be noted that the frequency instruction value ω1* may be calculated by employing a voltage detection value, instead of the voltage instruction value.
Based on expression (4), a slip calculator 17 employs the torque current instruction Iq* and the magnetic flux instruction Φ* to calculate an estimated slip speed value ωs^ for the motor. Further, in accordance with expression (5), a speed addition unit 18 calculates the estimated speed value ωr^ for the motor. It should be noted that in expression (4), T2 denotes a secondary time constant of the motor, and M denotes a mutual inductance of the motor.
ωs^=1/T2×M×Iq*/Φ* (4)
ωr^=ω1*−ωs^ (5 )
In this manner, the torque instruction value τ* is determined so that the estimated speed value ωr^ matches the speed instruction value ωr*. Then, a current is controlled so as to match the excited current instruction Id* and the torque current instruction Iq*, which are determined in accordance with the torque instruction value τ*. It should be noted that the torque instruction value τ* may be provided directly as an operating instruction, instead of being obtained through calculations based on a deviation between the speed instruction value ωr* and the estimated speed value ωr^.
Further, another method is disclosed in Japanese Patent No. 3070391 (paragraph [0011]), for example, whereby, when a large torque must be generated at the time a motor is activated, a magnetic flux instruction value is raised to obtain increased torque (∝ magnetic flux of a motor×a current).
Expressions (6) and (7) represent the relation between a d-axial magnetic flux Φd2 and a q-axial magnetic flux Φ2q of a motor. Iq and Id denote an actual torque current and an actual excited current, and an actual slip speed ωw is a difference between an inverter frequency ω1 and an actual speed ωr of a motor.
Φ2d=1/(1+T2·s)×(M×Id+T2×ωs×Φ2q) (6)
Φ2q=1/(1+T2·s)×(M×Iq+T2×ωs×Φ2d) (7)
In the above described vector control process, the actual slip speed ωs is appropriately controlled, so that d-axial magnetic flux Φ2d=M×Id and the q-axial magnetic flux Φ2q=0 are established. When the motor is not rotating, the actual slip speed ωs=inverter frequency ω1 is established. When the actual slip speed ωs/actual torque current Iq is increased, the q-axial magnetic flux Φ2q becomes smaller than 0, and as a result, the d-axial magnetic flux Φ2d is lowered and a desired torque may not be obtained.
Further, when there is an error in the constant of the motor in expression (1) or (2), accordingly, an error occurs in the frequency instruction value ω1* or the estimated speed value ωr^. When such an error occurs, the actual slip speed ωs becomes greater than the appropriate value, the magnetic flux is reduced, and a large torque may not be obtained. In addition, when a magnetic flux instruction value is increased, the magnetic flux of the motor is saturated, so that an actual magnetic flux may not be generated as instructed, and a large torque may not be output.
One objective of the present invention is to provide a motor driving apparatus that removes the adverse effects of the above described problems, and outputs a large, desired torque when a motor is activated.
To achieve this objective, according to the present invention, a motor driving apparatus comprises:
a current instruction calculator, for calculating a current instruction value based on a deviation between a speed instruction value and an estimated speed value;
a current controller, for controlling an output current based on the current instruction value;
a frequency instruction calculator, for calculating a frequency instruction value based on an output voltage instruction value or an output voltage detection value,
wherein, when an induction motor is to be activated, the current instruction value and the frequency instruction value are calculated directly using the speed instruction value, and an output current and an output frequency are controlled in accordance with the current instruction value and the frequency instruction value. Further, at the time of activation, the ratio of a frequency to a current is adjusted to a predetermined value, and slip speed/current is lower than after activation. The motor driving apparatus further comprises:
a slip calculator, for calculating slip speed for the induction motor based on either a torque current element of the current instruction value or a torque current detection value, and either a magnetic flux instruction value or an excited current instruction value,
wherein the slip speed for the induction motor that is activated is lower than a predetermined speed following the activation.
According to the present invention, when the induction motor is activated, the output current and the output frequency are adjusted in accordance with the speed instruction value, and the slip speed is adjusted to low, instead of the magnetic flux instruction value being increased. Thus, a reduction in the magnetic flux that is caused by excessive slip due to an error in the estimated speed can be prevented, and a desired torque can be output.
Other objects, features and advantages of the invention will become apparent from the following description of the embodiments of the invention taken in conjunction with the accompanying drawings.
The present invention will now be described in detail while referring to the accompanying drawings.
For a motor driving apparatus of a first embodiment of the present invention, different portions of the conventional example in
In this embodiment, a determination as to the time of activation, or after the activation, is performed to decide whether the speed instruction value ωr* has reached a value corresponding to a predetermined speed. However, time, for example, may be employed as a reference, and when a predetermined period of time has elapsed following the start of the operation of a driving apparatus, “after the activation” may be determined. Or when a torque corresponding value (=a voltage×a current/a speed) is raised to a predetermined value or greater, and is dropped again to its previous value, “after the activation” may be determined. At the time of activation, a frequency correction unit 33 corrects a frequency instruction value ω1*, so that there is zero deviation between the frequency instruction value ω1* and the speed instruction value ωr*. During this process, unlike expression (5), which is used for the conventional example, and without depending on the estimated speed value ωr^ or the estimated slip speed value ωs^, the frequency instruction value ω1* is calculated based on the speed instruction value ωr*, as shown in expression (8).
ω1*=ωr* (8)
Thus, in this embodiment, at the time of activation, the excited current instruction Id*, the torque current instruction Iq* and the frequency instruction ω1* are calculated in accordance with the speed instruction value ωr*, in order to remove the adverse effects of the frequency instruction value ω1* error and the estimated speed error, which conventionally occur during the calculation of the voltage instruction value or the voltage detection value. Thereafter, when the motor is started, and when the speed instruction value ωr* has reached a predetermined speed, e.g., 10% of the rated speed, the output of the speed control output correction unit 32 is returned from zero to equal that of the output value of the speed controller 11, and the output values of the torque instruction calculator 31 and the frequency correction unit 33 are reduced to zero. As a result, following the activation, as in the conventional example, the estimated speed value ωr^ and the estimated slip speed value ωs^ are employed to perform the control process. It should be noted that the output of the speed control output correction unit 32 may be adjusted to a small value, rather than to zero.
In this embodiment, at the time of activation, the actual slip speed ωs=frequency instruction value ω1*=speed instruction value ωr* is established. Further, since the flow of the actual torque current Iq is proportional to the actual slip speed ωs, both the frequency and the current are proportional to the speed instruction value ωr*. Therefore, the ratio of the frequency to the current at the time of activation is a predetermined value, e.g., a constant value. Furthermore, since the output of the torque instruction calculator 31 is adjusted at the time of activation, the value of the actual slip speed ωs, (the frequency instruction value ω1* at the time of activation)/the current, is smaller than the value after the activation. Therefore, magnetic flux reduction can be prevented, and a desired torque can be generated. It should be noted that in this embodiment I1=√(IqFB2+IdFB2)≈Iq is employed, in addition to the actual torque current Iq and the torque current detection value IqFB.
Furthermore, the speed instruction value ωr* is gradually increased from zero at a predetermined time change rate, so that a sharp increase in the actual slip speed ωs and the magnetic flux reduction are prevented, and a desired torque can be generated. At this time, the change rates for the frequency and the current should be proportional to the change rate for the speed instruction value ωr*. At the time of activation, the ratio of the changes for the frequency and the current is set to a predetermined value, i.e., a constant value, which should be smaller than that following the activation. For example, the speed instruction value ωr* is changed so it is approximately the equivalent of the rated slip speed, and the output current is changed so it is approximately the equivalent of the rated current. When the speed instruction value ωr* is increased, at a time change rate faster than (the rated slip speed/the secondary time constant of the motor)/10, until it reaches a predetermined value (e.g., about (the rated slip speed×a desired torque/the rated torque×2)), the torque is also increased, until it finally reaches a desired value.
As described above, according to this embodiment, at the time of activation, the excited current instruction Id*, the torque current instruction Iq* and the frequency instruction value ω1* are calculated in accordance with the speed instruction value ωr*, without being affected by the estimated slip speed value ωs^ and the estimated speed value ωr^. And the actual slip speed ωs (the frequency instruction value ω1* at the time of activation)/the current value can be adjusted to a small value, magnetic flux reduction can be reduced, and a desired torque can be generated.
While referring to
While referring to
ω1*=ωr*+ωs^×k (9)
In this embodiment, since k is zero at the time of activation, as in the first and second embodiments, the effect of the estimated slip speed value ωs^ can be removed, and an excited current instruction Id*, a torque current instruction Iq* and the frequency instruction value ω1* can be controlled in accordance with the speed instruction value ωr*. Further, at the time of activation, an actual slip speed ωs, (the frequency instruction value ω1* at the time of activation)/the current, can be adjusted until lower than after the motor has been activated, and a desired torque can be generated.
Furthermore, according to expression (4), the estimated slip speed value ωs^/(the torque current instruction Iq*/the magnetic flux instruction Φ*)=M/T2×k is established. This value is k times smaller at the time of activation (0≦k<1) than after the activation (k=1). Likewise, when the magnetic flux instruction is constant, the estimated slip speed value ωs^/the torque current instruction Iq* is k times (0≦k<1) smaller at the time of activation than after the activation. At this time, the torque current detection value IqFB may be employed, instead of the torque current instruction Iq*. In addition, since the actual torque current Iq is generally greater than the actual excited current Id, the actual torque current Iq may be regarded as current I1=√(IqFB2+IdFB2)≈Iq, and may be replaced by I1. Moreover, since the torque current instruction Iq* in expression (4) is proportional to the torque instruction τ*/the magnetic flux instruction Φ*, accordingly, the estimated slip speed value ωs^ is also proportional to the torque instruction τ*/(magnetic flux instruction Φ*2). Therefore, the estimated slip speed value ωs^/(the torque instruction τ*/(the magnetic flux instruction Φ*2)=a fixed value×k is established, and this value is k times (0≦k<1) smaller at the time of activation than after the activation (k=1). Similarly, when the magnetic flux instruction Φ*=M×excited current instruction Id* is defined, the estimated slip speed value ωs^/(the torque current instruction Iq*/the excited current instruction Id*) is k times smaller at the time of activation than after the activation. The same thing is applied when the excited current instruction Id* is replaced by the excited current detection value IdFB. The same effects are obtained by correcting a physical quantity used for slip calculation, instead of correcting the output of the slip calculator 17 in
In addition, when the speed instruction value ωr* is changed (e.g., for acceleration), it is assumed, based on expression (4), that the magnitude and the change rate of the actual torque current Iq are considerably greater than those of the actual excited current Id. Therefore, the change Δωs^ of the estimated slip speed value ωs^ is represented by expression (10). In expression (10), ΔIq* is replaced by output current change width ΔI.
Δωs^=1/T2×M×ΔI/Φ*×k (10)
When the change width of the speed instruction value ωr* is unchanged at the time of activation and after the activation, the difference in the change width of the frequency instruction value ω1* between at the time of activation and after the activation is equal to the difference in the change width of the estimated slip speed ωs^. Therefore, under a condition wherein the change width of the speed instruction value ωr* is equal, for example, to a condition wherein the change widths are regarded as equal so long as the difference in the two is within 1% of the rated speed, Δω^/ΔI is smaller by k at the time of activation than after the activation. That is, the ratio of the change of the output frequency to the change of the output current is smaller at the time of activation than after the activation.
As described above, the slip speed is increased less at the time of activation than after the activation. And as in the first and second embodiments, the magnetic flux reduction is reduced and a desired torque is generated. Further, as in the first embodiment, since the output of the frequency instruction calculator 20 is corrected based on a deviation between the speed instruction value ωr* and the output of the frequency instruction calculator 20, the same effects can be acquired as are obtained in the first embodiment. Further, when the slip speed correction unit 35 is provided to process the output of the slip calculator 17 in the second embodiment, the same effects can be acquired as are obtained in this embodiment.
While referring to
While referring to
While referring to
ωs=1/T2×M×Iq/Φ (11)
Further, the torque instruction τ* is gradually increased, and the ratio of the change of the frequency to the change of the current at the time of activation is adjusted to a predetermined value, i.e., a constant value. This value is smaller at the time of activation than after the activation. In this embodiment, as in the above embodiments, at the time of activation, the adverse effects of the frequency instruction ω1* error and the speed estimation error, which are conventionally caused by calculation of a voltage instruction or a voltage detection value, can be removed, and a desired torque can be generated. In this embodiment, the same effects can be obtained when, as in the fourth embodiment, the frequency instruction value ω1* is calculated by employing a voltage detection value.
While referring to
Since the frequency instruction value ω1*=the estimated slip speed value ωs^ is established, a slip speed correction unit 35 is adjusted to control the frequency instruction value ω1*. By adjusting the slip speed correction unit 35 at the time of activation, a frequency and a current can be controlled in accordance with operating instructions. As a result, at the time of activation, a predetermined ratio of the frequency to the current and a predetermined ratio of changes can be maintained. Therefore, as in the above embodiments, a desired torque can be generated without being affected by errors, such as a speed estimation error. After the motor has been activated, the output level of the frequency correction unit 33 is reduced, the torque instruction τ* is lowered, and the control process, as conventionally performed, is executed by employing the output of the speed controller 11 that is received from the speed control output correction unit 32.
It should be further understood by those skilled in the art that although the foregoing description has been made on embodiments of the invention, the invention is not limited thereto and various changes and modifications may be made without departing from the spirit of the invention and the scope of the appended claims.
Number | Date | Country | Kind |
---|---|---|---|
2005-049922 | Feb 2005 | JP | national |
Number | Name | Date | Kind |
---|---|---|---|
6909257 | Inazumi | Jun 2005 | B2 |
Number | Date | Country |
---|---|---|
3070391 | May 2000 | JP |
Number | Date | Country | |
---|---|---|---|
20060192521 A1 | Aug 2006 | US |