JP4198415B2 - Vector control inverter device - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to an inverter device that starts and drives a motor having a permanent magnet in a rotor by position sensorless vector control.
[0002]
[Prior art]
2. Description of the Related Art Household appliances having a rotating mechanism in recent years have been improved in performance such as improved usability and energy saving by adopting a motor having a permanent magnet as a rotor and driving it with an inverter device.
[0003]
Based on this trend, the inverter device employs a so-called vector control that separates the current into a magnetic flux axis direction component (d axis) of the permanent magnet motor and a torque direction component (q axis) orthogonal thereto and controls them independently. Tend to be. Furthermore, the adoption of a technique for driving without a position sensor by obtaining an induced voltage of a permanent magnet motor by calculation based on d-axis and q-axis currents is also expanding.
[0004]
This conventional position sensorless vector control technique will be described with reference to FIGS. The position sensorless vector control inverter device 1 shown in FIG. 7 includes a current control circuit 2, a rotor angular velocity estimation circuit 3, a current command angular velocity command formation circuit 4, and an integrator 5.
[0005]
The current control circuit 2 includes adders 21a and 21b, proportional integrators 22a and 22b, a coordinate converter 23, a PWM generator 24, a PWM inverter circuit 25, and a current detection circuit 26. The current detection circuit 26 includes current detectors 26a and 26b, a three-phase / two-phase converter 26c, and a vector rotator 26d. The angular velocity estimation circuit 3 includes an induced voltage estimation circuit 31 and a proportional integrator 32.
[0006]
Three-phase currents Iu, Iv and Iw (Iw is calculated from Iu and Iv) detected by current detectors 26a and 26b connected between the PWM inverter circuit 25 and the permanent magnet motor 6 are three-phase. / 2-phase converter 26c converts the two-phase currents Iα and Iβ equivalent thereto. The converted two-phase currents Iα and Iβ are further converted by the vector rotator 26d to obtain d-axis and q-axis component currents Id and Iq. In this conversion calculation, an estimated rotational position value θ described later is used. Here, the d-axis and the q-axis are rotational coordinate axes in which the direction of the magnetic flux produced by the permanent magnet of the rotor is the d-axis, and the direction orthogonal thereto is the q-axis.
[0007]
Deviations ΔId and ΔIq of the currents Id and Iq from the respective current command values Idr and Iqr are obtained by the adders 21a and 22b, and the output voltage command values Vd and Vq are obtained via the proportional integrators 22a and 22b. can get. The output voltage command values Vd and Vq are converted into values of a fixed biaxial coordinate system by the coordinate converter 23, and a three-phase pulse width modulation signal is formed by the PWM generator 24 based on this. A rotational position estimation value θ, which will be described later, is also used for the conversion calculation in the coordinate converter 23. The pulse width modulation signal is supplied to the PWM inverter circuit 25 and a voltage is applied to the armature of the motor 6. In this way, the current control circuit 2 supplies power to the motor 6, and the current depends on the current command values Idr and Iqr.
[0008]
Although there is a method of detecting the rotational position of the rotor necessary for calculation in the vector rotator 26d and the coordinate converter 23 by a rotation sensor such as an encoder attached to the motor 6, in recent years, for example, as shown in FIG. A position sensorless method estimated from the motor currents Id, Iq and the like is adopted.
[0009]
The induced voltage estimation circuit 31 receives the currents Id and Iq, the d-axis output voltage command value Vd, and the estimated angular velocity value ω of the rotor. Further, the induced voltage estimation circuit 31 stores the armature coil inductances Ld and Lq, and the resistance R, which are circuit constants of the motor 6.
[0010]
The induced voltage estimation circuit 31 uses these input values and circuit constants to estimate the d-axis direction of the induced voltage generated in the armature coil by the magnetic flux generated by the permanent magnet (d-axis recognized by the inverter device 1). The direction component (Ed) is calculated by the following equation.
Ed = Vd−R · Id−Ld · p · Id + ω · Lq · Iq (1)
Here, p is a differential operator.
[0011]
The obtained induced voltage estimated value Ed is input to the proportional integrator 32 and a value calculated by the following equation is output as the angular velocity estimated value ω.
ω = -G1 · Ed-G2 · ∫Ed · dt Equation (2)
Here, G1 and G2 are gain constants.
The obtained angular velocity estimated value ω is input to the current command angular velocity command forming circuit 4.
[0012]
During steady operation, the estimated angular velocity value ω is output as it is as the angular velocity command value ωr to the integrator 5. The integrator 5 integrates the angular velocity estimated value ω according to the following equation (3), and outputs the rotor rotational position estimated value θ.
θ = ∫ωr · dt = ω · dt Equation (3)
The obtained rotational position estimation value θ is used for calculation in the vector rotator 26d and the coordinate converter 23.
[0013]
Based on the induced voltage estimated value Ed calculated by the equation (1), the proportional integrator 32 performs the adjusting operation according to the equation (2), so that the induced voltage estimated value Ed converges to “zero” in a short time. To do. When the d-axis induced voltage estimated value Ed converges to zero, the d-axis recognized (estimated) by the inverter device 1 coincides with the direction of the magnetic flux created by the permanent magnet, and the rotational position estimated value θ is an actual value of the motor 6. It is equal to the rotor position and the angular velocity estimate ω is equal to the actual angular velocity of the rotor. In this way, the rotational position and angular velocity of the rotor are detected without using a position sensor.
[0014]
The current command angular velocity command forming circuit 4 also outputs a d-axis current command value Idr and a q-axis current command value Iqr. Since the d-axis current Id is a component that does not contribute to the torque generation of the motor 6, a “zero” value is normally output as its command value Idr. The adder 21a calculates the deviation ΔId between the d-axis current command value Idr and the detected value Id. When the proportional integrator 22a performs an adjusting operation for converging the deviation ΔId to “zero”, the current Id becomes equal to the command value Idr.
[0015]
On the other hand, the current Iq is a component that contributes to the torque generation of the motor 6, and the command value Iqr is formed by the current command angular velocity forming circuit 4 and output according to the desired operating state of the load. The adder 21b calculates the deviation ΔIq of the q-axis current command value Iqr and the detected value Iq. When the proportional integrator 22b performs an adjusting operation for converging the deviation ΔIq to “zero”, the current Iq becomes equal to the command value Iqr.
[0016]
The above arithmetic processing excluding the PWM inverter circuit 25 is periodically processed by an arithmetic unit such as a DSP (Digital Signal Processor). The calculation includes, for example, a three-phase / two-phase converter 26c, a vector rotator 26d, an induced voltage estimation circuit 31, a proportional integrator 32, a current command angular velocity command forming circuit 4, an integrator 5, adders 21a and 21b, a proportional integral. The units 22a and 22b, the coordinate converter 23, and the PWM generator 24 are executed in this order.
Thus, the currents Id and Iq of the motor 6 are controlled by the vector control inverter device 1 so as to coincide with the command values Idr and Iqr without using a position sensor.
[0017]
[Problems to be solved by the invention]
Next, a conventional starting method when the motor 6 is started by the vector control inverter device 1 of FIG. 7 will be described. Then, in the explanation, the problems of the conventional starting method and the problem to be solved by the present invention are clarified.
[0018]
FIG. 8 shows the state of switching of the current command values Idr and Iqr and the angular velocity command value ωr to the integrator 5 from the start to the steady operation stage, step by step.
[0019]
FIG. 9 is a vector diagram of the current and magnetic flux of the motor 6 at each stage, and FIG. 10 is a time chart showing changes in current and angular velocity. The uppermost curve (A) in FIG. 10 shows the transition of the ideal angular velocity ωx of the motor 6 at the start. In FIG. 8 (and FIGS. 2 and 5 described later), the step is represented by “S”.
[0020]
After starting, first (1) a positioning step is performed. At this stage, as shown in step S31 of FIG. 8, the current command value Iqr and the angular velocity command value ωr are set to zero, and the d-axis current command value Idr is increased from zero to a predetermined value Ids in a short time. Here, the value of the current Ids is set to a value that generates sufficient torque to position the rotor of the motor 6 in the d-axis direction assumed by the inverter device 1 against the frictional torque of the load. As a result, the magnetic flux direction created by the permanent magnet and the magnetic flux direction created by the armature current are substantially matched.
[0021]
(1) of FIG. 9 shows the relationship between the current vector i, the magnetic flux vector φ of the permanent magnet, and the respective axes in such a positioned state. In the vector diagram of FIG. 9 (and FIG. 3 to be described later), the d-axis and q-axis assumed by the inverter device 1 are di and qi, respectively, and the actual d-axis determined from the magnetic flux direction created by the permanent magnet of the rotor. , Q-axis are represented by dm and qm, respectively.
[0022]
In the state where the positioning of (1) is completed, the direction of the current vector i and the direction of the magnetic flux vector φ of the permanent magnet coincide with each other, and the dm axis and the di axis substantially coincide. Since the current vector i and the magnetic flux vector φ are in phase, no rotational torque is generated. The estimated rotational position value θ remains constant because the angular velocity command value ωr is zero. Therefore, the di-axis, qi-axis, dm-axis, qm-axis, and armature current vector i are all stationary when viewed from the fixed coordinate side. (1) in FIG. 10 shows changes with time in current and angular velocity at this stage. In FIG. 10 (and FIGS. 4 and 6 described later), the angular velocity ωm represents the actual angular velocity of the rotor, and ωi represents the angular velocity of the current vector i.
[0023]
After positioning in step S31, the process proceeds to (2) forced rotation stage in step S32. At this stage, the angular velocity command value ωr is gradually increased from zero to a predetermined value ωs. When the rotational position command value θ obtained by integrating the angular velocity command value ωr (the angular velocity of the motor 6 is controlled in an open loop at this stage, θ is not an estimated value but a command value) starts to increase, the di-qi axis Starts rotation at angular velocity ωr. As a result, as shown in the vector diagram of (2) in FIG. 9, the di-qi axis and the dm-qm axis are separated, and a phase difference is generated. Since the current vector i also starts to rotate at the angular velocity ωr with the direction of the di-axis, a phase difference is generated between the current vector i and the magnetic flux vector φ of the permanent magnet. The torque T acting on the rotor is a value calculated by the outer product of the current vector i and the magnetic flux vector φ. Accordingly, a rotational torque T is generated by the generation of the phase difference, and the rotor starts to rotate so as to follow the current vector i. The phase difference between the current vector i and the magnetic flux vector φ is determined according to the generated torque T and the load torque. Note that vm in FIG. 9 (and FIG. 3 described later) indicates an induced voltage vector induced in the armature coil by the rotation of the permanent magnet. The (2) forced rotation stage in FIG. 10 shows changes with time in current and angular velocity at this stage.
[0024]
When the angular velocity command value ωr reaches the predetermined value ωs, the process proceeds to (3) current switching stage of step S33. In this (3) current switching stage, while maintaining the angular velocity command value ωr at the predetermined value ωs, the current command value Idr is changed from Ids to zero, and Iqr is changed from zero to Iqs at the angular velocity λ according to the following equation. That is, the currents Id and Iq are switched.
Idr = Ids · cos (λ · t) (4)
Iqr = Iqs · sin (λ · t) Equation (5)
Here, λ is a constant, and t is (3) the elapsed time from the start of the current switching stage.
[0025]
Since the current command values Idr and Iqr start to change according to the above equation, the angular velocity of the current vector i is (3) the current switching angular velocity λ is added to the angular velocity command value ωr at the moment when the current switching stage is started. Switch to (ωs + λ). Since the rotor immediately after switching is subjected to a sufficiently large rotational torque T determined by the current vector i having the absolute value Id and the magnetic flux vector φ, the angular velocity ωm of the rotor follows the current vector i and follows the angular velocity. It changes rapidly to (ωs + λ). That is, as shown in part (3) (a) of FIG. 10, a discontinuous change in angular velocity occurs in the rotor at the switching timing. This discontinuous change in angular velocity causes the motor 6 to generate vibration and noise.
[0026]
The phase difference between the current vector i and the magnetic flux vector φ spreads to an angle at which the load torque corresponding to the new angular velocity (ωs + λ) and the generated torque T become equal, and rotates at the angular velocity (ωs + λ) while maintaining the phase difference. . On the other hand, since the angular velocity of the di-qi axis is maintained at ωs, the phase of the di-qi axis continues to be delayed at the angular velocity λ when viewed from the dm-qm coordinates rotating at the angular velocity (ωs + λ). Finally, the qi axis and the current vector i coincide with each other. The vector diagrams in the (3) current switching stage and its end state are shown in (2-3) and (3) in FIG. 9, and the state of change in current, angular velocity, etc. is shown in (3) in FIG. The current Iqs, which is the absolute value of the current vector i in this end state, is also set to a value that can generate a torque sufficient to accelerate the motor 6.
[0027]
When step S33 ends, the process proceeds to the (4) angular velocity switching stage of step S34. From this (4) angular velocity switching stage, sensorless rotor angular velocity and rotational position detection calculation begins to function, and current vector control starts based on the estimated (detected) rotor position (detected dm-qm axis). Is done. As the angular velocity command value ωr, the estimated angular velocity value ω estimated by the angular velocity estimation circuit 3 is output as it is. That is, the angular velocity command value ωr becomes equal to the angular velocity estimation (detection) value ω from this stage. The values of the current command values Idr and Iqr are maintained at zero and Iqs, which are values at the end of the previous stage, respectively.
[0028]
When the current switching is completed, λ = 0, and the angular velocity ωi of the current vector i instantaneously decreases by λ. At the same time, by setting ωr = ω, the sensorless angular velocity and rotational position detection calculation described in the section of “Prior Art” starts to function, and the angular velocity command value ωr tries to converge to the angular velocity ωm of the rotor. .
[0029]
Although the angular velocity ωm of the rotor tends to decrease due to the decrease in the angular velocity ωi of the current vector i by λ, the angular velocity ωm does not rapidly decrease because the rotor has inertia. Therefore, the inverter 1 starts adjusting the angular velocity estimated value ω so that the d-axis induced voltage estimated value Ed becomes zero when the angular velocity ωm of the rotor is slightly lower than (ωs + λ) (that is, the angular velocity command The value ωr begins to converge to the actual rotor angular velocity ωm).
[0030]
The convergence of the angular velocity command value ωr to the rotor angular velocity ωm means that the di-qi axis coincides with the dm-qm axis, and the current vector i having only the qi component coincides with the qm axis. As a result, the phase difference between the current vector i and the magnetic flux vector spreads to π / 2. The value of the torque T acting on the rotor due to the spread of the phase difference suddenly increases simultaneously with the convergence of ωr to ωm.
The rapid increase in the rotational torque T and the accompanying rapid increase in the rotor angular velocity ωm cause the motor 6 to generate vibrations and noise (the state shown in FIG. 10 (A)). A vector diagram in the middle of convergence of the angular velocity command value ω to the rotor angular velocity ωm is shown in (3-4) of FIG. 9, and a vector diagram of the converged state is shown in (4) of FIG.
[0031]
After the angular velocity command value ωr converges to the actual angular velocity ωm of the rotor, the rotor continues to have a torque T determined by the outer product of the current vector i of the absolute value Ids flowing in the qi axis direction and the magnetic flux vector φ in the qm axis direction. Work. Since this generated torque T has a phase difference of π / 2, it is larger than (3) torque T in the current switching stage. The rotor is accelerated and the angular velocity ωm continues to rise until it matches the load torque. FIG. 10 (4) shows how the angular velocity and the current value change in the (4) acceleration switching stage.
[0032]
When the angular velocity reaches the steady state, the process proceeds to (5) steady operation stage of step S35. At this stage, the rotational speed control is performed so that the angular velocity command value ωr equal to the estimated angular velocity value ω matches the angular velocity target value ωc (not shown) given from the outside. The current command angular velocity command forming circuit 4 outputs the q-axis current command value Iqr by, for example, the following proportional integration calculation.
Iqr = G3 · (ωc-ω) + G4 · ∫ (ωc-ω) · dt Equation (6)
Here, G3 and G4 are constants. Since the current command value Idr does not contribute to torque generation, for example, zero output is continued. FIG. 10 (5) shows how the angular velocity and current change during the (5) steady operation stage.
[0033]
As described above, in the conventional starting method, a rapid change occurs in the actual angular velocity ωm of the rotor in the portions shown in FIGS. 10A and 10B, which generates vibration and noise in the motor 6. It is a cause.
[0034]
The present invention has been made in view of such circumstances, and an object of the present invention is to control a sudden change in the actual angular velocity of the rotor and to start a motor without generating vibration and noise. To provide an apparatus.
[0035]
[Means for Solving the Problems]
In order to achieve the above object, the vector control inverter device according to claim 1 is configured such that a current of a permanent magnet motor having a permanent magnet provided on a rotor is an estimated value (Ed) of a magnetic flux axial direction component of an induced voltage by the permanent magnet. A vector-controlled inverter device that separates and controls the magnetic flux axis (di-axis) direction component estimated by the calculation of zero and the torque axis (qi-axis) direction component orthogonal thereto The forced rotation stage in which the current vector of the current to be supplied to the motor is rotated in the di-axis direction and rotated at the angular speed ωs together with the di-axis to rotate the rotor, and the di-axis angular speed is once set to zero, and the current vector at the angular speed ωs a current switching step of rotating in the qi-axis direction, and then rotating the rotor with an estimated angular velocity value ω based on the estimated magnetic flux axis direction component (Ed) of the induced voltage. It is characterized by moving to a stage.
[0036]
By adopting such a configuration, a rapid change in the rotor angular velocity ωm at the moment of shifting from the forced rotation stage to the current switching stage of the di-axis current and the qi-axis current can be suppressed, and motor vibration and noise can be suppressed. effective.
[0037]
The vector control inverter device according to claim 2 is the vector control inverter device according to claim 1, wherein the estimated value (Ed) of the magnetic flux axis direction component of the induced voltage is zero after the current switching stage. At this point, the rotational speed control is performed in a state where the magnetic flux axis direction component estimated value (Ed) is kept at zero.
[0038]
By adopting such a configuration, in addition to the effect of claim 1, the continuity of the angular velocity ωi of the current vector i at the moment of switching the rotor angular velocity control from the open loop control to the closed loop control is maintained. There is an effect that a sudden change does not occur in the angular velocity ωm of the rotor, and motor vibration and noise are suppressed.
[0039]
The vector control inverter device according to claim 3 is the vector control inverter device according to claim 1, wherein the estimated value (Ed) of the magnetic flux axis direction component of the induced voltage is zero after the current switching stage. At this point, the shift to the rotational speed control in which the magnetic flux axis direction component estimated value (Ed) is kept at zero is made, and the qi-axis current is changed until the di-axis current becomes zero. It is characterized by maintaining.
[0040]
By adopting such a configuration, in addition to the effect of claim 2, the continuity of torque acting on the rotor at the moment of switching the rotor angular speed control from the open loop control to the closed loop control is maintained. There is an effect that a sudden change does not occur in the angular velocity ωm of the child, and motor vibration and noise are suppressed.
[0041]
The vector control inverter device according to claim 4 estimated the current of a permanent magnet motor having a permanent magnet provided on the rotor by a calculation in which the magnetic flux axial direction component estimated value (Ed) of the induced voltage by the permanent magnet is zero. Magnetic flux axis (di-axis) direction component and torque axis (qi-axis) direction perpendicular thereto component Is a vector control inverter device that is controlled separately and independently, and the current vector of the current supplied to the permanent magnet motor is rotated in accordance with the external angular velocity command value ωa together with the di axis while being fixed in the di axis direction. After passing through a forced rotation stage for rotating the child and a current switching stage in which the di-axis angular velocity is once set to zero and the current vector is rotated in the qi-axis direction according to the external angular velocity command value ωa, the magnetic flux axis of the induced voltage is passed. The present invention is characterized by moving to a stage where the rotor is rotated by the angular velocity estimated value ω based on the direction component estimated value (Ed).
[0042]
With this configuration, in addition to being able to increase the rotational speed in accordance with the external angular velocity command value ωa, the rotor angular velocity ωm at the moment of shifting from the forced rotation stage to the current switching stage of the di-axis current and the qi-axis current. Abrupt changes can be suppressed, and motor vibration and noise can be suppressed.
[0043]
The vector control inverter device according to claim 5 is the vector control inverter device according to claim 4, wherein after the current switching stage according to claim 4, the flux axial direction component estimated value of the induced voltage is passed. When (Ed) becomes zero, the rotational speed control is performed in a state where the estimated magnetic flux axis direction component estimated value (Ed) is kept at zero.
[0044]
With this configuration, in addition to the effect of claim 4, the continuity of the angular velocity ωi of the current vector i is maintained at the moment of switching the rotor angular velocity control from the open loop control to the closed loop control. There is an effect that a sudden change in the angular velocity ωm of the motor does not occur and the generation of motor vibration and noise can be suppressed.
[0045]
The vector control inverter device according to claim 6 is the vector control inverter device according to claim 4, wherein after the current switching stage according to claim 4, the estimated value of the magnetic flux axial direction component of the induced voltage is obtained. When (Ed) becomes zero, the control shifts to the rotational speed control in which the magnetic flux axial direction component estimated value (Ed) is kept at zero, and after the shift, the qi-axis current is converted into the rotor angular velocity ωm The adjustment is performed so as to match the external angular velocity command value ωa.
[0046]
With this configuration, in addition to the effect of the fourth aspect, the continuity of the torque acting on the rotor at the moment of switching the rotor angular velocity control from the open loop control to the closed loop control is maintained. There is an effect that a sudden change in the angular velocity ωm does not occur, and motor vibration and noise are suppressed.
[0049]
DETAILED DESCRIPTION OF THE INVENTION
Hereinafter, an embodiment of the present invention will be described with reference to the drawings. Since this embodiment has many parts that are the same as or similar to the conventional embodiments described in detail in the sections of “Prior Art” and “Problems to be Solved by the Invention”, the description of overlapping parts is omitted. Only the differences will be described in detail. Moreover, in the following figures, the same code | symbol is attached | subjected to the part which is the same as that of a prior art, or is equivalent.
[0050]
(First embodiment)
FIG. 1 is an electrical configuration diagram of the inverter device of the present embodiment, FIG. 2 is a flowchart showing a switching state of a current command value, an angular velocity command value, etc. from the start to the steady operation stage, and FIG. FIG. 4 is a time chart of electric current, angular velocity and the like.
[0051]
The electrical configuration of the inverter device shown in FIG. 1 is almost the same as FIG. 7 showing the prior art. The only difference is that the d-axis induced voltage estimated value Ed, which is the calculation result of the induced voltage estimating circuit 31, is additionally input to the current command angular velocity command forming circuit 4. Since the operation of each component circuit, converter, etc. other than the current command angular velocity command forming circuit 4 is the same as that described in the section “Prior Art”, description thereof will not be repeated. The operation of the current command acceleration command forming circuit 4 will be clarified in the description of the activation sequence according to the flow of FIG.
[0052]
First, the (1) positioning stage of Step S1 shown in FIG. The operation at this stage is the same as the (1) positioning stage in step S31 of FIG. In the rotor, a magnetic flux vector φ formed by a permanent magnet is positioned in the di-axis direction, which is the d-axis assumed (recognized) by the inverter device 1. A vector diagram of the positioned state is shown in (1) of FIG. 3, and a change with time of the current command value and the like at this stage is shown in (1) of FIG. Note that the value of the current Ids is set to a value that generates a torque sufficient to accelerate the motor 6.
[0053]
Subsequently, the process proceeds to (2) forced rotation stage of step S2. The operation at this stage is also the same as the operation at (2) forced rotation stage in step S32 in FIG. A vector diagram at this stage is shown in (2) of FIG. 3, and changes with time of the current command value and the like are shown in (2) of FIG. At the end of this stage, the angular velocity command value ωr that is the input of the integrator 5 and the actual angular velocity ωm of the motor 6 are both equal to the predetermined value ωs. In other words, the armature current vector i has an absolute value Ids and is directed in the di-axis direction and is rotated at an angular velocity ωs. The magnetic flux vector φ of the rotor is oriented in the dm-axis direction and is rotated at the same angular velocity ωs with a constant phase delay from the current vector i. The phase difference between the current vector i and the magnetic flux vector φ is determined by the condition that the generated torque T (the outer product of the current vector i and the magnetic flux vector φ) and the load torque are equal. A vector vm in the vector diagram of FIG. 3 indicates an induced voltage vector generated in the armature coil by the rotation of the magnetic flux vector φ, the direction always coincides with the qm axis, and the absolute value is the actual value of the rotor. Is proportional to the angular velocity ωm.
[0054]
When step S2 ends, the process proceeds to the current switching stage. This current switching stage is composed of three steps, ie, (3a) current switching stage in step S3, (3b) current switching stage in step S4, and (3c) current switching stage in step S5. At this stage, the d-axis current command value Idr is increased from Ids to zero, and the q-axis current command value Iqr is increased from zero to Iqs. That is, switching between the d-axis current Id and the q-axis current Iq is performed.
[0055]
With the start of the current switching stage (3a) in step S3, current command values Idr and Iqr for the d-axis and q-axis are output from the current command angular velocity command forming circuit 4 according to the following calculation formula.
Idr = Ids · cos (ωs · t) Equation (7)
Iqr = Iqs · sin (ωs · t) Equation (8)
Here, ωs is (3a) the value of the angular velocity command value ωr immediately before moving to the current switching stage, and t is (3a) the elapsed time from the start of the current switching stage. These formulas are different from formulas (4) and (5) for calculating the current command value in the step (3) current switching stage of step S33 of the prior art in that the angular velocity λ in the formula is replaced by ωs. It is.
[0056]
On the other hand, the value of the angular velocity command value ωr is maintained at ωs in the prior art, whereas zero is output in the present embodiment. Since the angular velocity command value ωr is zero, the rotation of the di-qi axis is stopped. However, since the current command value for each axis is output according to equations (7) and (8), the current vector i rotates at the angular velocity ωs. That is, (3a) In the current switching stage, the angular velocity ωi of the current vector i is maintained at the angular velocity ωs immediately before switching. Accordingly, the rotor also rotates at the angular velocity ωs immediately before switching in a form that follows this. A vector diagram in this state is shown in (3a) of FIG. 3, and changes in angular velocity and the like are shown in (3a) of FIG. The portion (a) of FIG. 10 of the prior art corresponds to the portion (a) of FIG. 4, and in the case of this embodiment, there is no sudden change in the angular velocity ωm of the rotor. Therefore, the occurrence of vibration and noise at the time of switching, which was a problem in the prior art, is eliminated in this embodiment.
[0057]
In this (3a) current switching stage, the value of the d-axis induced voltage estimated value Ed is simultaneously monitored by the current command angular velocity command forming circuit 4. The induced voltage vector vm that coincides with the qm axis continues to rotate at the angular velocity ωs together with the dm-qm axis, while the di-qi axis stops. Therefore, the estimated value Ed of the di-axis direction component of the induced voltage vector vm gradually decreases and finally becomes zero.
[0058]
The stage where the value of Ed becomes zero is the current switching stage (3b) in step S4. A vector diagram in this state is shown in (3b) of FIG. 3, and states of the current command value and the like are shown in (3b) of FIG. The value of the q-axis current command value when this stage is reached is stored as Iqk.
[0059]
Subsequently, the process proceeds to (3c) current switching stage of step S5. At this stage, the d-axis current command value Idr continues to decrease according to the equation (7). On the other hand, the q-axis current command value is maintained at the previously stored value of Iqk.
Simultaneously with the start of this (3c) current switching stage, the angular velocity estimated value ω estimated by the angular velocity estimating circuit 3 is output as it is as the angular velocity command value ωr. That is, the estimation calculation of the angular velocity ωm of the rotor starts to function from this stage. (3c) Since the rotor is rotating at the angular velocity ωs at the moment of shifting to the current switching stage, the estimated angular velocity value ω converges to ωs by the estimation calculation, and the value ωs is output as the angular velocity command value ωr. Is done. Thereafter, since the estimated value ω of the rotor angular velocity ωm (almost equal to ωm) is output as the angular velocity command value ωr, the di-qi axis continues to rotate with the dm-qm axis almost coincident. A vector diagram of this state is shown in (3c) of FIG.
[0060]
The generated torque T is proportional to the outer product of the magnetic flux vector φ and the current vector i, that is, the product of the magnetic flux vector φ and the qm-axis component of the current vector i. Here, the qi-axis current command value Iqr is maintained at a constant value Iqk, and the qi-axis and the dm-axis almost coincide with each other, so that the generated torque T is maintained at a constant value through the current switching stage (3c). . Accordingly, the angular velocity ωm of the rotor changes at a constant value ωs during this stage. The above transition is shown in (3c) of FIG.
[0061]
As shown in the vector diagram of FIG. 9 (3), the di-qi axis and the dm-qm axis are used for switching from the (3) current switching stage in FIG. 10 to the (4) acceleration switching stage in FIG. The angular velocity command value ωr was switched to the estimated value ω in a state where they did not match. Therefore, as a result of the angular velocity estimation calculation, a rapid rotation of the di-qi axis in the dm-qm axis direction and a corresponding rapid rotation of the current vector i in the qm axis direction occur, as shown in FIG. A sudden change in the current vector i and the rotor angular velocity ωm occurred.
[0062]
On the other hand, in the present embodiment, the angular velocity command value ωr is switched to the estimated value ω in a state where the di-qi axis and the dm-qm axis coincide with each other as shown in (3b) of FIG. It is. Therefore, the current vector i and the rotor angular velocity ωm do not change as shown in the portion (a) in FIG. 4 (4), and no vibration, noise, etc. occur at this switching portion.
[0063]
When the d-axis current command value Idr in step S5 becomes zero, the process proceeds to the next (4) steady operation stage of step S6. The operation at this stage is the same as the operation in the step (5) steady operation stage of step S35 in FIG. The rotational speed control is performed so that the angular velocity command value ωr equal to the estimated angular velocity value ω matches the angular velocity target value ωc (not shown) given from the outside. The current command angular velocity command forming means 4 outputs the q-axis current command value Iqr, for example, according to the above-described equation (6). FIG. 4 (4) shows how the angular velocity and current change during the (4) steady operation stage.
[0064]
As is clear from the above description, according to the present embodiment, the rapid change in the current vector i and the rotor angular velocity ωm seen in the portions (a) and (b) of FIG. Since it disappears as shown in the parts (a) and (b) of 4, the generation of motor vibration and noise in this part is eliminated.
[0065]
(Second Embodiment)
In the first embodiment, the rotational speed of the motor 6 is increased along the ideal angular speed curve assumed by the current command angular speed command forming circuit 4 until the angular speed ωc in the steady operation stage is reached after starting. Therefore, the first embodiment is an embodiment that is convenient when the steady-state angular velocity ωc is given stepwise from the outside. On the other hand, the second embodiment is convenient when the angular velocity curve at the time of activation is given from the outside, and the angular velocity command value ωa starts from zero and draws a gentle curve to reach the steady-state angular velocity. It is a starting method.
[0066]
The operation of the second embodiment has many parts that are similar to those of the first embodiment. Therefore, the same operation part is only described to that effect, the duplicate description is omitted, and only different points will be described.
FIG. 5 shows a flow chart of start-up in the second embodiment, and FIG. 6 shows changes with time of current command values and the like. The electrical configuration and vector diagram of the inverter device are the same as those in FIGS. 1 and 3 in the case of the first embodiment.
[0067]
When the start is started, the (1) positioning step of step S21 in FIG. 5 is executed. This stage is the same as the operation in the first embodiment. FIG. 6 (1) shows the transition of the current command value and the like, and FIG. 3 (1) shows a vector diagram of the state in which the rotor is positioned.
[0068]
When positioning is completed, the process proceeds to (2) forced rotation stage of step S22. From this stage, an external angular velocity command value ωa is given to the current command angular velocity command forming circuit 4 (not shown). FIG. 6A shows an example of a curve of the external angular velocity command value ωa. In the first embodiment, the angular velocity command value ωr increases from zero to the predetermined value ωs according to the gradient generated by the current command angular velocity command forming circuit 4, but in this embodiment, according to the curve of the angular velocity command value ωa from the outside. Increase to a predetermined value ωs. The vector diagram at this stage is shown in (2) of FIG. 3, and the transition of the angular velocity command value and the like is shown in (2) of FIG.
[0069]
When the external angular velocity command value ωa reaches the predetermined value ωs, the process proceeds to (3a) current switching step in step S23. The operation at this stage is almost the same as step S3 of the first embodiment, but the point that the current command value is switched at the angular velocity ωa as in the following equation is (7), (8) of the first embodiment. Different from formula.
Idr = Ids · cos (ωa · t) Equation (9)
Iqr = Iqs · sin (ωa · t) (10)
The vector diagram at this stage is shown in (3a) of FIG. 3, and the transition of the angular velocity and the like is shown in (3a) of FIG.
[0070]
As the value of the angular velocity command value ωr, zero is output as in the first embodiment, and the current vector i rotates at the angular velocity ωa. (3a) Since the actual angular velocity ωm of the rotor immediately before the current switching stage is ωs, the current vector i and the angular velocity ωm of the rotor are changed at the part (a) of FIG. Does not occur. Therefore, no vibration or noise occurs in this portion.
[0071]
(3a) In the current switching stage, the d-axis induced voltage estimated value Ed is monitored as in the case of the first embodiment. Then, when the value of Ed becomes zero, the process proceeds to (3b) current switching stage of step S24. This state differs from the first embodiment in that the angular velocity ωm of the rotor is ωs in the first embodiment, whereas in this embodiment, it is equal to the external angular velocity command value ωa at that time. It is a point. A vector diagram at this stage is shown in (3b) of FIG. 3, and angular velocity and the like are shown in (3b) of FIG.
[0072]
Subsequently, the process proceeds to (3c) current switching stage of step S25. From this stage, as in the first embodiment, the value of ω obtained by estimating the angular velocity ωm of the rotor is output as it is as the angular velocity command value ωr. That is, the angular velocity estimation calculation starts to function. The d-axis current command value Idr continues to decrease toward zero according to the equation (9).
[0073]
As the q-axis current command value Iqr, the current command angular velocity command forming circuit 4 performs, for example, the following proportional integral calculation so that the angular velocity command value ωr equal to the estimated angular velocity value ω matches the external angular velocity command value ωa. Is output.
Iqr = G5 · (ωa-ω) + G6 · ∫ (ωa-ω) · dt Equation (11)
Here, G5 and G6 are constants. However, (3b) so that there is no change in the value of the current command value Iqr before and after the current switching stage, that is, (3c) the value of Iqr immediately after moving to the current switching stage is (3b) The initial value of the integral value of the second term of equation (11) is set so as to be equal to the value. F (ωa, ω) in steps S25 and S26 in FIG. 5 means the expression on the right side of the expression (11). By outputting the current command value Iqr in this way, the rotational speed control is performed so that the actual angular velocity ωm of the rotor matches the angular velocity command value ωa given from the outside.
[0074]
The vector diagram at (3c) current switching stage is shown in (3c) of FIG. 3, and the transition of the current command value and the like is shown in (3c) of FIG. In FIG. 6 (a) of the present embodiment, which corresponds to the portion (b) of FIG. 4 of the prior art, changes in the current vector i and the angular velocity ωm of the rotor are the same as in the first embodiment. Does not occur. Therefore, vibration and noise do not occur at this switching portion.
[0075]
The stage where the current command value Idr continues to decrease and becomes zero is the (3d) current switching stage of step S26. The vector diagram of this stage is shown in (4) of FIG. 3, the angular velocity is shown in FIG. Shown in (3d).
[0076]
Then, it moves to (4) steady operation stage of step S27. At this stage, since the d-axis current command value Id does not contribute to torque generation, a zero value is normally output. On the other hand, the q-axis current is continuously controlled according to the equation (11). Thus, the rotational speed of the rotor is controlled so that the angular speed ωm of the rotor coincides with the external angular speed command value ωa.
[0077]
As is clear from the above description, according to the second embodiment, the rapid change in the current vector i and the rotor angular velocity ωm seen in the portions (a) and (b) of FIG. However, as shown in the parts (a) and (b) of FIG. 6, the motor vibration and noise are eliminated in this part.
[0078]
【The invention's effect】
According to the present invention, sudden fluctuations in torque acting on the rotor are suppressed in the starting stage until the motor reaches a steady operation state from a stopped state. Therefore, since a rapid change in the rotor angular velocity does not occur, a smooth motor start with less vibration and noise can be realized.
[Brief description of the drawings]
FIG. 1 is an electrical configuration diagram showing an embodiment of a vector control inverter device of the present invention.
FIG. 2 is a flowchart showing a start sequence according to the first embodiment.
FIG. 3 is a vector diagram of various electrical quantities at each stage of startup according to the first embodiment.
FIG. 4 is a time chart showing a transition of a current command value and the like at each stage of start of the first embodiment.
FIG. 5 is a view corresponding to FIG. 2 for the second embodiment.
FIG. 6 is a view corresponding to FIG. 4 for the second embodiment.
FIG. 7 is a view corresponding to FIG.
FIG. 8 is a view corresponding to FIG.
FIG. 9 is a view corresponding to FIG.
FIG. 10 is a view corresponding to FIG. 4 showing the prior art.
[Explanation of symbols]
In the figure, 1 is a vector control inverter device, 2 is a current control circuit, 3 is an angular velocity estimation circuit, 4 is a current command angular velocity command formation circuit, 5 is an integrator, 6 is a permanent magnet motor, 31 is an induced voltage estimation circuit, di Is the estimated magnetic flux axis, qi is the estimated torque axis, Id is the di-axis current, Iq is the qi-axis current, Idr is the di-axis current command value, Iqr is the qi-axis current command value, Ed is the magnetic flux axis of the induced voltage (di Axis) direction component estimated value, ω is the estimated angular velocity of the rotor, ωs is a predetermined angular velocity, ωr is the angular velocity command value, ωa is the external angular velocity command value, ωm is the actual angular velocity of the rotor, i is the current vector, vm Represents an induced voltage vector, and φ represents a magnetic flux vector of the permanent magnet.

Claims (6)

The current of a permanent magnet motor provided with a permanent magnet in the rotor is orthogonal to the magnetic flux axis (di-axis) direction component estimated by the calculation that the estimated value (Ed) of the magnetic flux axis direction component of the induced voltage by the permanent magnet is zero. A vector-controlled inverter device that is independently controlled separately from the torque axis (qi axis) direction component,
A forced rotation stage in which the current vector of the current supplied to the permanent magnet motor is rotated in the di-axis direction at an angular velocity ωs while the current vector is fixed in the di-axis direction, and the rotor is rotated.
a current switching stage in which the di-axis angular velocity is once set to zero and the current vector is rotated in the qi-axis direction at an angular velocity ωs;
Is passed, the process proceeds to the stage of rotating the rotor with the estimated angular velocity value ω based on the estimated magnetic flux axial direction component (Ed) of the induced voltage.   At the time when the estimated magnetic flux axis direction component estimated value (Ed) of the induced voltage becomes zero during the current switching stage, the rotor is moved at the angular velocity estimated value ω based on the estimated magnetic flux axis direction component (Ed). The vector control inverter device according to claim 1, wherein the vector control inverter device moves to a rotation stage.   At the time when the estimated magnetic flux axis direction component estimated value (Ed) of the induced voltage becomes zero during the current switching stage, the rotor is moved at the angular velocity estimated value ω based on the estimated magnetic flux axis direction component (Ed). 2. The vector controlled inverter device according to claim 1, wherein the qi-axis current maintains a current value at the time of transition until the di-axis current becomes zero while shifting to the rotation stage. The current of a permanent magnet motor provided with a permanent magnet in the rotor is orthogonal to the magnetic flux axis (di-axis) direction component estimated by the calculation that the estimated value (Ed) of the magnetic flux axis direction component of the induced voltage by the permanent magnet is zero. A vector-controlled inverter device that is independently controlled separately from the torque axis (qi axis) direction component ,
A forced rotation stage in which a current vector of a current supplied to the permanent magnet motor is rotated in accordance with the external angular velocity command value ωa together with the di-axis while the current vector is fixed in the di-axis direction;
a current switching stage in which the di-axis angular velocity is once set to zero and the current vector is rotated in the qi-axis direction according to the external angular velocity command value ωa;
Is passed, the process proceeds to the stage of rotating the rotor with the estimated angular velocity value ω based on the estimated magnetic flux axial direction component (Ed) of the induced voltage.   An estimated angular velocity based on the estimated magnetic flux direction component (Ed) when the estimated magnetic flux direction component estimated value (Ed) of the induced voltage becomes zero during the current switching step according to claim 4. The vector control inverter device according to claim 4, wherein the process proceeds to a stage of rotating the rotor by ω.   An angular velocity estimation based on the magnetic flux axis direction component estimated value (Ed) when the magnetic flux axis direction component estimated value (Ed) of the induced voltage becomes zero after the current switching stage according to claim 4 has passed. 5. The process proceeds to a stage of rotating the rotor at a value ω, and after the transition, the qi-axis current is adjusted so that the rotor angular speed ωm matches the external angular speed command value ωa. Vector control inverter device.

Images