JP2016178814A - Motor control device and electric apparatus - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a motor control device capable of appropriately compensating for a pulsation component of torque in accordance with a state of a motor.SOLUTION: Regarding a motor (5) that is rotationally driven by an inverter (3), the motor control device comprises: an axial error estimation part (15) for estimating an axial error (Δθ) that is a difference between an estimated rotation angle and an actual rotation angle; a torque controller (22) which outputs a torque command correction value (I) that increases/decreases a torque command value (I) for suppressing the axial error (Δθ); and a driving signal generation part (14) that outputs a driving signal (PWM signal) to the inverter (3), based on the increased/decreased torque command value (I). The torque controller (22) includes: a correction value limitation part (227) which limits a variation range of the torque command correction value (I) so as to be settled within a designated limitation range; and a limitation range setting part (228) for increasing/decreasing the limitation range, based on a state of the motor (5).SELECTED DRAWING: Figure 1

Description

  The present invention relates to an electric motor control device and an electric device.

  As a background art of this technical field, the abstract of the following Patent Document 1 states that “when a load device that is a rotational drive target of an AC synchronous motor generates a periodic disturbance, the periodic disturbance is suppressed. ”Provide an electric motor control device that reduces input power”, “Extract 21 the pulsation component of torque generated by the load device, and limit the current component that corrects the pulsation component in the torque control 22 that compensates for it. It can be achieved by providing a limiter 227. "

JP 2006-180605 A

However, Patent Document 1 does not particularly disclose the limit process for limiting the current component for correcting the pulsation component, in that the strength of the limit process is changed according to the situation. If the same limit processing is performed regardless of the state of the motor, it may not be possible to suppress the torque disturbance because the necessary correction is not given when the torque disturbance is likely to occur, such as when the motor starts or when the speed changes suddenly. Arise. Further, in a situation where a strong limit process can be applied, if the limit process remains weak, a problem arises in that excessive correction current flows and power consumption increases.
The present invention has been made in view of the above-described circumstances, and an object thereof is to provide an electric motor control device and an electric device that can appropriately compensate for a pulsating component of torque according to the state of the electric motor.

In order to solve the above problems, in the motor control device of the present invention,
An axis error estimation unit that estimates an axis error that is a difference between an assumed rotation angle and an actual rotation angle for an electric motor that is rotationally driven by an inverter;
In order to suppress the shaft error, a torque controller that outputs a torque command correction value that increases or decreases the torque command value and a drive signal generation that outputs a drive signal to the inverter based on the increased or decreased torque command value The torque controller comprises:
A correction value limiting unit that limits the fluctuation range of the torque command correction value so as to be within a specified limit range;
And a limit range setting unit that increases or decreases the limit range based on the state of the electric motor.

  According to the motor control device and the electric apparatus of the present invention, it is possible to appropriately compensate the torque pulsation component according to the state of the motor.

1 is a block diagram of an electric motor control system according to a first embodiment of the present invention. It is a vector diagram explaining a control axis in the first embodiment. It is an internal block diagram of the Δθ estimator in the first embodiment. It is a block diagram explaining the principle until the axis | shaft error generation | occurrence | production in 1st Embodiment. It is a wave form diagram of each part in a 1st embodiment. It is a block diagram explaining the estimation principle of the pulsation torque component in 1st Embodiment. FIG. 3 is a block diagram of a ΔT _m estimator in the first embodiment. It is a wave form diagram of each part at the time of acceleration of a 1st embodiment. It is a wave form chart of each part at the time of deceleration of a 1st embodiment. It is a wave form chart of each part at the time of acceleration of a 2nd embodiment. It is a wave form chart of each part at the time of deceleration of a 2nd embodiment. It is a wave form chart of each part at the time of acceleration of a 3rd embodiment. It is a wave form diagram of each part at the time of the deceleration of 3rd Embodiment. It is a wave form diagram of each part at the time of acceleration / deceleration of 4th Embodiment.

[First Embodiment]
<Overall Configuration of First Embodiment>
Next, the configuration of the motor control system according to the first embodiment of the present invention will be described with reference to the block diagram shown in FIG.
The rotational speed command generator 1 outputs a rotational speed command ω _r ^* which is a target value of the rotational speed (mechanical angular frequency) of the PM motor (permanent magnet type synchronous motor) 5. The DC power supply 4 includes a diode bridge 42 that rectifies an AC voltage supplied from an AC power supply 41 that is a commercial power supply, and a smoothing capacitor 43 that suppresses a pulsating component included in the output voltage of the diode bridge 42. Yes. The DC voltage output from the smoothing capacitor 43 is V0.

The inverter 3 performs pulse width modulation on the DC voltage supplied from the DC power supply 4 based on the supplied PWM (pulse width modulation) signal, and applies it to the PM motor 5. The PM motor 5 drives a compression mechanism 6 that is a load. The current detector 7 detects the current I 0 supplied from the DC power supply 4 to the inverter 3 and supplies it to the controller (motor control device) 2. The controller 2 calculates the voltage to be applied to the PM motor 5 so that the rotation speed of the PM motor 5 gradually matches the rotation speed command ω _r ^* , and converts the PWM signal based on the calculation result into an inverter 3 is supplied. The induced voltage detector 24 detects the induced voltage of the PM motor 5. Specifically, in any phase of the PM motor 5, the induced voltage is measured by measuring the terminal voltage of that phase at the timing when the DC voltage V0 is not applied.

When the number of poles of the PM motor 5 is P, the number of pole pairs is P / 2. Inside the controller 2, the proportional device 19 outputs an electrical angular frequency command ω ₁ ^* by multiplying the rotational speed command ω _r ^* by the number of pole pairs P / 2. The current reproducer 8 reproduces the three-phase alternating currents I _u , I _v and I _w supplied to the PM motor 5 by calculation based on the detection result of the current I 0. For this reproduction method, for example, the method described in Japanese Patent Application Laid-Open No. 8-19263 can be used. The reproduced results are called reproduced three-phase alternating currents I _uc , I _vc , I _wc . The coordinate converter 9 converts the three-phase AC currents I _uc , I _vc , and I _wc on the three-phase axis into the currents I _dc and I _qc on the control axis.

Here, the control axis will be described with reference to the vector diagram shown in FIG. In this embodiment, the control axis is an axis on a rotational coordinate that rotates at an electrical angular frequency, and the direction of the magnetic flux of the actual permanent magnet in the PM motor 5 is the d axis, and the axis orthogonal to this is the q axis. To do. Further, the d-axis and q-axis assumed in the controller 2 are referred to as a dc-axis and a qc-axis. As shown in FIG. 2, there is an axis error Δθ between the d-axis and q-axis and the dc-axis and qc-axis. The above-described currents I _dc and I _qc are currents on the dc axis and the qc axis. Note that the q-axis current I _q actually generated in the q-axis direction is a current that controls the torque of the PM motor 5 and is sometimes referred to as “torque current”.

Returning to FIG. 1, the I _q0 ^* generator 10 _gives a slight delay element to the current I _{qc and outputs} the result as a q-axis current command base value I _q0 ^* . The adder 16 adds the q-axis current command base value I _q0 ^* and a q-axis current command correction value I _qSIN ^* , which will be described later, and _outputs the addition result as the q-axis current command value I _q ^* . The I _d ^* generator 11 generates a d-axis current command value I _d ^* . If the PM motor rotor is a non-salient pole type, the d-axis current command value I _d ^* is normally zero.

Based on the d-axis and q-axis current command values I _d ^* and I _q ^* and the electrical angular frequency command ω ₁ ^* , the voltage command calculator 12 calculates the d-axis and q-axis voltage command values V _d ^* and V _q. ^{* Is} output. The dq inverse converter 13 performs coordinate conversion of the d-axis and q-axis voltage command values V _d ^* and V _q ^* and converts them into three-phase AC voltage commands V _u ^* , V _v ^* , and V _w ^* . The PWM pulse generator 14 outputs a PWM signal for switching the inverter 3 based on these three-phase AC voltage commands V _u ^* , V _v ^* , V _w ^* .

The Δθ estimator 15 outputs an estimated axis error Δθ _c that is an estimated value of the axis error Δθ shown in FIG. Details thereof will be described later. The zero generator 17 outputs a command value (zero value) for the estimated axis error Δθ _c . The subtractor 31 outputs a deviation (−Δθ _c ) by subtracting the estimated axis error Δθ _c from the command value (zero value). The proportional compensator 18 multiplies the deviation (−Δθ _c ) by a predetermined gain K _PLL, and outputs the result as an electrical angular frequency correction value Δω ₁ . The adder 32 adds the electrical angular frequency command ω ₁ ^* and the electrical angular frequency correction value Δω ₁ and outputs the result as the estimated electrical angular frequency ω _1c . The integrator 20 integrates the estimated electrical angular frequency ω _1c to calculate a d-axis angle estimated value θ _dc that is an estimated value of the d-axis angle θ _d and supplies it to the coordinate converter 9 and the dq inverse converter 13. To do.

In this way, by adding the electrical angular frequency command ω ₁ ^* and the electrical angular frequency correction value Δω ₁ by the adder 32, feedback control is realized such that the estimated axis error Δθ _c approaches a zero value. As shown in FIG. 2, when the axial error Δθ is a positive value, the dc axis and qc axis are ahead of the d axis and q axis, so the estimated electrical angular frequency ω _{1c is set} to the electrical angular frequency command. By lowering than ω ₁ ^*, the increase of the d-axis angle estimated value θ _dc becomes moderate, and the axis error Δθ can be reduced. Conversely, when the axis error Δθ is a negative value, the increase of the electrical angular frequency command ω ₁ ^* increases the d-axis angle estimated value θ _dc faster, and the axis error Δθ can be reduced. .

In this embodiment, if the estimated axis error Δθ _c output from the Δθ estimator 15 is a positive value, the deviation (−Δθ _c ) output from the subtractor 31 and the electrical angle output from the proportional compensator 18. The frequency correction value Δω ₁ is a negative value. On the other hand, if the estimated axis error Δθ _c is a negative value, the electrical angular frequency correction value Δω ₁ becomes a positive value, so that control that reduces the axis error Δθ is possible. Thus, in the present embodiment, the estimated d-axis angle value θ _dc inside the controller is used as the actual magnetic pole position in the PM motor, that is, the d-axis angle without using a position sensor that detects the magnetic pole position of the PM motor 5. can be brought close to theta _d, it can be realized the position sensorless control.

The ΔT _m estimator 21 calculates an estimated torque pulsation component ΔT _mc that is an estimated value of the torque pulsation component ΔT _m based on the estimated axis error Δθ _c . Here, the torque pulsation component ΔT _m is a difference between the load torque T _L generated by the compression mechanism 6 and the motor torque T _m generated by the PM motor 5. The torque controller 22 outputs a q-axis current command correction value I _qSIN ^* based on the estimated torque pulsation component ΔT _mc . Details of the ΔT _m estimator 21 and the torque controller 22 will be described later.

<Detailed configuration of each part>
(Δθ estimator 15)
Next, the details of the Δθ estimator 15 will be described with reference to the block diagram shown in FIG.
In the Δθ estimator 15, the subtractor 152 subtracts the current I _qc from the q-axis current command value I _q ^* . The proportional device 151 obtains the estimated axis error Δθ _c by multiplying the subtraction result by a predetermined constant K0. The current I _qc varies when the load torque varies and a deviation occurs between the actual d-axis angle θ _d and the estimated d-axis angle θ _dc . Therefore, on the contrary, the axis error Δθ can be estimated based on the movement of the current I _qc . The configuration of FIG. 3 simulates this. When it is necessary to further determine the precise axis estimated error [Delta] [theta] _c, for example, may be calculated estimated axis error [Delta] [theta] _c according to equation (3) or the like in JP 2002-272194.

(ΔT _m estimator 21)
Next, the details of the ΔT _m estimator will be described. First, the principle of generating the axis error Δθ will be described with reference to FIG. FIG. 4 is a block diagram of a transfer function until an axis error Δθ is generated by the q-axis voltage V _q .
In the subtractor 36 of FIG. 4, the voltage disturbance V _D is subtracted from the q-axis voltage V _q to obtain the voltage applied to the PM motor 5. When the resistance value of the winding of the PM motor 5 is R and the inductance is L, the transfer function for deriving the q-axis current I _q from this voltage is (1 / (R + sL)). Note that s represents a differential operator used for Laplace transform. The transfer function for deriving the motor torque T _m from the q-axis current I _q is (P / 2) · (3K _e / 2) = 3P · K _e / 4. Here, P is the number of poles of the PM motor 5, and _Ke is a power generation constant.

The subtracter 37, by which the load torque T _L from the motor torque T _m is subtracted, the torque ripple component [Delta] T _m obtained. The integrator 51 divides the torque pulsation component ΔT _m by the inertia J of the rotor and integrates it to output the rotational speed ω _r of the PM motor 5. The proportional device 52 outputs the electrical angular frequency ω ₁ by multiplying the rotational speed ω _r by the number of pole pairs P / 2. The integrator 53 outputs a d-axis angle θ _d by integrating the electrical angular frequency ω ₁ . The subtractor 38 outputs an axis error Δθ by subtracting the d-axis angle θ _d from the estimated d-axis angle value θ _dc .

Here, an example of the voltage disturbance V _D will be described. For example, when the magnet magnetic flux of the PM motor 5 is not uniform and there is a variation in magnetization, or when there is a variation in phase between windings, these are equivalently affected as a periodic voltage disturbance V _D. In addition, disturbance due to the influence of the arm short circuit prevention period (dead time) in the inverter 3 also occurs as a voltage disturbance V _{D having a} frequency 6 times the drive frequency of the inverter. As the compression mechanism 6, for example, a reciprocating compression mechanism, a single rotary compression mechanism, a twin rotary compression mechanism, a two-stage compression rotary compression mechanism, or the like can be considered. In a reciprocating compression mechanism, a rotary compression mechanism, and the like, the load torque _TL varies greatly with one rotation of the electric motor as one cycle.

Here, the waveform diagrams of the respective parts in FIG. 4 are shown in FIGS. These diagrams are waveform diagrams in the case where the load torque T _L includes a component that oscillates in a sinusoidal shape at the angular frequency ω _d . The angular frequency ω _d is determined by the structure of the compression mechanism 6, but is generally the same as the mechanical angular frequency ω of the PM motor 5 or an integer multiple thereof. In FIG. 5A, the average value of the motor torque _{Tm and} the average value of the load torque _TL coincide in the long term. Therefore, as shown in FIG. 5B, the torque pulsation component ΔT _m which is the difference between the two is only the vibration component. FIG. 5C shows the vibration component Δω _r included in the rotational speed ω _r .

The vibration component Δω _r is obtained by integrating the torque pulsation component ΔT _m and has a waveform whose phase is delayed by 90 degrees compared to ΔT _m . The magnitude of the vibration of the vibration component Δω _r itself varies depending on the inertia J, but the phase may be considered to be delayed by approximately 90 degrees. Since the axial error Δθ is obtained by further integrating the vibration component Δω _r and inverting the sign (the sign is inverted due to the definition shown in FIG. 2), the phase is advanced by 90 degrees (integral Since the sign is reversed after 90 degrees delay, it is advanced 90 degrees). The axis error Δθ is shown in FIG. Comparing FIGS. 5B and 5D, it can be seen that the torque pulsation component ΔT _m and Δθ are observed as vibration waveforms having the same phase.

Next, the principle of obtaining the torque pulsation component ΔT _m from the axis error Δθ based on the relationship shown in FIG. 5 will be described with reference to the block diagram shown in FIG. FIG. 6A shows a transfer function from the torque pulsation component ΔT _m to the axis error Δθ in the integrator 51, the proportional unit 52, the integrator 53, and the subtractor 38 shown in FIG. The transfer function from the axis error Δθ to the torque pulsation component ΔT _m can be obtained by inversely converting the block diagram of FIG. The result is shown in FIG. 6B is summarized as shown in FIG. 6C.

According to FIG. 6 (c), theoretically, the estimated axis error [Delta] [theta] _c may be determined estimated torque ripple component [Delta] T _mc by differentiating upstairs. However, it is practically difficult to second-order differentiate the estimated axis error Δθ _c . Since the estimated axis error Δθ _c is originally an estimated value and includes a large amount of noise or the like of the detected value, there is a problem that the estimation error increases when using differentiation. In addition, performing second-order differentiation is limited by the calculation cycle. Therefore, paying attention to the point that “the torque pulsation component ΔT _m and other disturbance components are periodic functions”, s = jω _d can be substituted into the transfer function of FIG.

Then, as shown in FIG. 6D, a value obtained by multiplying the axis error Δθ by a constant (2Jω _d ² / P) can be estimated as the torque pulsation component ΔT _m . This result agrees with the waveform relationship of FIGS. 5 (b) and 5 (d). A configuration embodying FIG. 6D is a ΔT _m estimator 21 shown in the block diagram of FIG. Inside the ΔT _m estimator 21, the proportionalizer 211 multiplies the estimated axis error Δθ _c by a constant “2J / P”. The multiplier 212 squares an estimated angular frequency ω _dc that is an estimated value of the angular frequency ω _d .

Here, if the angular frequency ω _d is the same as the mechanical angular frequency ω of the PM motor 5, the estimated angular frequency ω _dc is a value obtained by multiplying the estimated electrical angular frequency ω _1c by the reciprocal (2 / P) of the number of pole pairs. 2ω _1c / P). When a speed reducer is inserted between the PM motor 5 and the compression mechanism 6, the result obtained by further multiplying the multiplication result (2ω _1c / P) by the reduction ratio may be used as the estimated angular frequency ω _dc. . Multiplier 213 multiplies the output signals of proportional multiplier 211 and multiplier 212 and outputs the result as estimated torque pulsation component ΔT _mc .

(Torque controller 22)
Returning to FIG. 1, the configuration of the torque controller 22 will be described.
Inside the torque controller 22, the single phase-dq coordinate converter 223 decomposes the estimated torque pulsation component ΔT _mc into a sine component ΔT _qs and a cosine component ΔT _ds according to the following equation (1).

ΔT _ds = ΔT _mc · cos ω _dc t
ΔT _qs = −ΔT _mc · sin ω _dc t (1)

According to Expression (1), if the estimated torque pulsation component ΔT _mc includes a frequency component of the estimated angular frequency ω _dc , the average value of the cosine component ΔT _ds and the sine component ΔT _qs is “0” according to the amount. It becomes a value that is not. However, since the cosine component ΔT _ds and the sine component ΔT _qs contain many frequency components that are twice the estimated angular frequency ω _dc , it is desirable to remove these frequency components. Therefore, the first-order lag filters 224d and 224q set the time constant T _ATR so as to remove frequency components more than twice the estimated angular frequency ω _dc , and these frequency components are obtained from the cosine component ΔT _ds and the sine component ΔT _qs. Remove.

Therefore, the output signals of the first-order lag filters 224d and 224q coincide with the cosine component and sine component of the estimated angular frequency ω _dc included in the estimated torque pulsation component ΔT _mc . The zero generator 35 outputs zero values that are target values of the cosine component and the sine component. The subtractors 33 and 34 subtract the output signals of the first-order lag filters 224d and 224q from zero values, respectively. The integrators 225d and 225q multiply and integrate the output signals of the subtracters 33 and 34 by a predetermined constant K _iATR , respectively.

The limiter unit 227 limits the output signals of the integrators 225d and 225q, and outputs the result as current command correction values I _ds and I _qs . The limit value setting unit 228 sets a limit value for the limiter unit 227. The dq-single-phase inverse converter 226 inversely converts the current command correction values I _ds and I _qs into a single-phase signal according to the following equation (2), and outputs the result as a q-axis current command correction value I _qSIN ^* . .

I _qSIN ^* = I _ds · cos ω _dc t−I _qs · sin ω _dc t (2)

As described above, the q-axis current command correction value I _qSIN ^* is added to the q-axis current command base value I _q0 ^* by the adder 16, thereby increasing or decreasing the q-axis current command value I _q ^* . Thereby, the axis error Δθ and the estimated axis error Δθ _c are controlled so as to approach zero values. When the estimated axis error Δθ _c approaches the zero value, the estimated torque pulsation component ΔT _mc proportional to this approaches the zero value. Since the pulsation component ΔT _mc becomes a direct current amount after the coordinate conversion is performed by the equation (1), the integrators 225d and 225q can eliminate the deviation. That is, the torque controller 22 is equivalent to a compensation element whose gain is infinite at the angular frequency ω _d when viewed from the outside.

In limit value setting unit 228, load torque T _L is obtained in order to limit current command correction values I _ds and I _qs . Therefore, how to determine the load torque _TL in the limit value setting unit 228 will be described. The load torque T _L can be calculated by the theoretical formula shown in the following formula (3).

T _L = (3/2) P _n {K _e I _qfb + (L _d −L _q ) I _dfb I _q } (3)

In Equation (3), P _n is the number of pole pairs (= P / 2), K _e is the power generation constant, L _d and L _q are d-axis and q-axis inductances, I _dfb and I _qfb are d-axis and q-axis current observation values Represents. Here, assuming that the d-axis and q-axis inductances L _d and L _q are equal and the q-axis current observation value I _qfb is equal to the current I _qc output from the coordinate converter 9, Equation (3) is simplified. The following equation (4) is obtained.

T _L ≈ (3P · K _e / 4) I _qc (4)

Therefore, the limit value setting unit 228 sequentially calculates the load torque T _{L according} to the formula (3) or (4).

<Operation of First Embodiment>
(Operation during acceleration)
Next, details of the operation of the present embodiment, particularly the operation of limiting the current command correction values I _ds and I _qs by the limiter unit 227 and the limit value setting unit 228 will be described.
FIGS. 8A to 8D are waveform diagrams showing examples of waveforms at various parts during acceleration of the electric motor control system of the present embodiment. FIG. 8A shows a rotation speed command ω _r ^* and a rotation speed command. A waveform example of the speed ω _r is shown. The rotational speed ω _r is obtained by multiplying the estimated electrical angular frequency ω _1c output from the adder 32 (see FIG. 1) by the reciprocal “2 / P” of the number of pole pairs. A state in which the PM motor 5 is stably driven at a constant speed is referred to as a “steady state”.

The steady state speed determination value Δω _rth shown in FIG. _8A is a value for determining whether or not the PM motor 5 has reached the steady state, and the rotational speed ω _r is ω _r ^* ± Δω _rth . If it is within the range, the limit value setting unit 228 determines that the PM motor 5 has reached a steady state by satisfying other conditions. In the example of FIG. 8A, the rotational speed ω _r is lower than the rotational speed command ω _r ^{* at} time t0, and therefore the rotational speed ω _r increases after time t0. The rotational speed ω _r reaches the range of ω _r ^* ± Δω _rth at time t3, and is substantially equal to the rotational speed command ω _r ^* at time t5 thereafter.

FIG. 8B shows a waveform example of the load torque T _L and the calculation load torque T _LC calculated by the limit value setting unit 228. The load torque T _L is obtained by the equation (3) or (4) as described above, and this varies sequentially due to various factors. On the other hand, the calculation load torque T _LC is a value used for calculation, and is set to a value equal to the load torque T _L at the start of the torque review period T ₁ shown in FIG. the torque review period T ₁ is kept constant until the end.

FIG. 8C shows a waveform example of the estimated axis error Δθ _c and a steady state axis error determination value ± Δθ _th . The steady state axis error determination value ± Δθ _th is one index for determining the above-described “steady state”, and if the fluctuation range of the estimated axis error Δθ _c falls within the range of the steady state axis error determination value ± Δθ _th , By satisfying other conditions, the limit value setting unit 228 determines that the PM motor 5 has reached a steady state.

FIG. 8D shows the q-axis current command correction value I _qSIN ^* and the compensation amount limit value ± I _th . As described above, the limit value setting unit 228 sets limit values for the current command correction values I _ds and I _qs that are values on the dq coordinates, and these limit values are set on the values on the single-phase coordinates. The compensation amount limit value ± I _{th shown} in FIG.

In the present embodiment, in order to calculate the compensation amount limit value I _th , values such as the induced voltage change rate K _v and the q-axis current command average value I _qave ^* will be described. The rotational speed ω _r increases during the period from time t0 to time t5 in FIG. 8A. When the rotational speed ω _r increases, the induced voltage detected by the induced voltage detector 24 is also proportional to the rotational speed ω _r . Then increase. The induced voltage measured at the start of the rotational speed change (time t0 in the example of FIG. 8A) and the induced voltage measured when the rotational speed command is reached (time t5 in the example of FIG. 8A). The ratio of the brazing induced voltage to the predicted induced voltage obtained by calculation is referred to as “induced voltage change rate K _v ”.

Further, the q-axis current command average value I _qave ^* indicates an average value obtained by removing a pulsating component from the torque current, that is, the q-axis current command value I _q ^* . Note that the q-axis current command base value I _q0 ^* output from the I _q0 ^* generator 10 may actually be used as the q-axis current command average value I _qave ^* . This is the same in the operation at the time of deceleration, which will be described later, and in other embodiments. At (time t0 in the example in FIG. 8 (a)) when the PM motor 5 starts to accelerate, the limit value setting unit 228, and the induced voltage variation rate K _v, and a q-axis current command average value I _qave ^* And the product “K _v · I _qave ^* ” of the two is set to a provisional compensation amount limit value I _th (that is, an initial value of the compensation amount limit value I _th ).

As described above, during acceleration, the limit value setting unit 228 updates the calculation load torque T _LC every torque review cycle T ₁ (set to a value equal to the load torque T _L at that time). At this time, immediately before the calculation load torque T _LC is updated, the limit value setting unit 228 compares the load torque T _L with the calculation load torque T _LC . When the comparison result is “T _LC <T _L ”, it is desirable to increase the torque current, that is, the q-axis current command value I _q ^* from the current level, so that the compensation amount limit value I _th is increased from the current level. It is desirable.

Therefore, the limit value setting unit 228, if the comparison result is "T _LC <T _L", to the current compensation amount limit value I _th, the increment correction corresponding to the deviation (T _L -T _LC) performed, to the result as a new compensation amount limit value I _th. For example, when the proportional constant and K _i, K _i and added value of (T _L -T _LC) with respect to the previous compensation amount limit value I _th, or when a new compensation amount limit value I _th. On the other hand, if the comparison result is "T _LC ≧ T _L" is previous compensation amount limit value I _th is maintained.

Referring to FIG. 8B again, at the times t3 and t4, the calculation load torque _TLC before the update is smaller than the load torque _TL . Thus, at time t3, t4 of FIG. 8 (d) compensation amount limit value I _th is increased stepwise. Further, at time t4 to t7, since the relationship of "T _LC ≧ T _L" is maintained, the compensation amount limit value I _th in the same period is maintained at a constant value. In the limit value setting section 228, until a steady state of the PM motor 5 is detected, the compensation amount limit value I _th is determined as described above.

Here, the “steady state” described above will be described in more detail. The limit value setting unit 228
(1) Speed condition: The rotational speed ω _r is in the range of ω _r ^* ± Δω _rth .
(2) Axis error condition: The estimated axis error Δθ _c is within the range of the steady state axis error determination value ± Δθ _th .
(3) Time condition: A period in which the speed condition and the axis error condition are satisfied continues for a predetermined steady state determination time T _j .
When the three conditions are met, the state of the PM motor 5 is determined to be the “steady state”.

According to FIG. 8A, the speed condition is satisfied after time t3. Further, according to FIG. 8C, the axis error condition is satisfied at time t6. Since all of the speed condition, the axis error condition, and the time condition are satisfied at the time t7 when the steady state determination time _Tj has elapsed from the time t6, at the time t7, the limit value setting unit 228 indicates that “PM motor 5 is stationary. It is determined that the state has been reached.

In the steady state, it is considered that the compensation amount limit value I _th until then is over-compensated. _Therefore , in the limit value setting unit 228, the compensation amount limit value I _th is changed to the q-axis current command average value I _qave ^* . Is done. In the example of FIG. 8D, since the steady state is established at time t7, the compensation amount limit value I _th is set to the q-axis current command average value I _qave ^* and is lower than the previous value. Yes. In the example of FIG. 8D, the limit value is constant for ease of explanation. However, since the q-axis current command average value I _qave ^* actually changes, the compensation amount in the steady state according to this change. The limit value I _th also changes. This is the same in the operation at the time of deceleration, which will be described later, and in other embodiments.

If the speed condition or the axis error condition is not established after the steady state is established, the steady state is not established again. Here, the mode in which the steady state is not established is
(Aspect 1: Acceleration) When the rotational speed command ω _r ^* is higher than the rotational speed ω _r by a predetermined value or more,
(Aspect 2: Deceleration) When the rotational speed command ω _r ^* is lower than the rotational speed ω _r by a predetermined value or more,
(Aspect 3: Large axis error) When the estimated axis error Δθ _c exceeds the steady state axis error determination value ± Δθ _th ,
There are three possible modes.

First, for “mode 1: acceleration”, the same operation as the operation after time t0 described above is repeated. The “mode 2: deceleration” will be described in detail later as “operation during deceleration”. Here, “Aspect 3: Large axial error” will be described. One of the reasons why the estimated axis error Δθ _c increases is that the load torque T _L varies due to various non-periodic disturbances.

At time t8 in FIG. 8C, the steady state is not established because the estimated axis error Δθ _c starts to fluctuate beyond the steady state axis error determination value ± Δθ _th . In this case, the limit value setting unit 228 returns the compensation amount limit value I _th to the value immediately before the steady state is established (in the example of FIG. 8D, immediately before time t7). As a result, strong compensation is again applied to the axis error Δθ. Thereafter, as described above, when the three conditions of the speed condition, the axis error condition, and the time condition are met, the limit value setting unit 228 repeats the operation in the steady state again.

(Operation during deceleration)
Next, with reference to FIGS. 9A to 9D, an operation example during deceleration in the present embodiment will be described. FIGS. 9A to 9D are waveform diagrams showing examples of waveforms at various parts during deceleration of the motor control system of the present embodiment. In the example of FIG. 9A, since the rotational speed ω _r is higher than the rotational speed command ω _r ^{* at} time t10, the rotational speed ω _r decreases after time t10. Then, the rotational speed ω _r eventually reaches the range of ω _r ^* ± Δω _rth , and is substantially equal to the rotational speed command ω _r ^* at time t11 thereafter.

In FIG. 9D, at the time t10 when the deceleration is started, the q-axis current command average value I _qave ^* at that time is set to the compensation amount limit value I _th . Thereafter, the compensation amount limit value I _th may be gradually decreased. However, in the limit value setting unit 228 of the present embodiment, until a steady state is established in preparation for an unexpected change in the load torque T _L. The compensation amount limit value I _th is kept constant.

The limit value setting unit 228 sequentially determines whether or not a steady state has been established, as in the above-described acceleration (FIGS. 8A to 8D). In FIG. 9A, since the rotational speed ω _r has reached ω _r ^* ± Δω _rth before time t11, the speed condition is satisfied before time t11. In FIG. 9C, since the estimated axis error Δθ _c falls within the range of the steady state axis error determination value ± Δθ _th at time t14, the axis error condition is satisfied at time t14. Then, the steady state is established at time t16 when the steady state determination time T _j has elapsed from time t14.

When the limit value setting unit 228 determines that the steady state is established, the compensation amount limit value I _th is changed to the q-axis current command average value I _qave ^* as in the case of the acceleration described above. The Accordingly, at time t16, the compensation amount limit value I _th is lower than the previous value. After the steady state is established at time t16, the steady state may not be established again. The modes are “Aspect 1: Acceleration”, “Aspect 2: Deceleration”, and “Aspect 3: Large axial error”. In any case, the limit value setting unit 228 executes the operation already described.

That is, for “Aspect 1: Acceleration”, the same operation as the operation after time t0 in FIGS. 8A to 8D is repeated. For “mode 2: deceleration”, the same operation as the operation after time t10 in FIGS. 9A to 9D is repeated. As for the “mode 3: large axis error”, the compensation amount limit value I _th is returned to the value just before the steady state is established (just before time t16 in the example of FIG. 9D) as described above. It is. After time t18 in FIGS. 9A to 9D, an example of a waveform when “mode 3: large axis error” occurs is shown.

As described above, in the present embodiment, the case PM motor 5 is in a steady state is set low amount of compensation limit value I _th, unsteady state (especially aspects 1: acceleration and aspects 3: axial error Univ.) in so set high compensation amount limit value I _th, the q-axis current command correction value I _QSIN ^* range of variation in the non-steady state can be wider than in the steady state. Thereby, in the unsteady state, the situation where the behavior becomes unstable due to the disturbance of the load torque T _L or the like can be effectively suppressed. Further, when the PM motor 5 is in a steady state, the power consumption can be reduced by reducing the compensation amount limit value I _th , thereby suppressing the pulsation of the DC voltage V 0 output from the DC power supply 4. There is also an effect that can be done.

[Second Embodiment]
(Operation during acceleration)
Next, a second embodiment of the present invention will be described.
The configuration of the second embodiment is the same as that of the first embodiment (FIGS. 1 to 7), but the setting contents of the compensation amount limit value I _th in the unsteady state are different. The contents will be described with reference to FIGS. FIGS. 10A to 10D are waveform diagrams showing examples of waveforms at various parts during acceleration of the motor control system of the present embodiment.

In the example of FIG. 10A, since acceleration starts at time t20 and the rotational speed ω _r is lower than the rotational speed command ω _r ^* , the rotational speed ω _r increases after time t20. In FIG. 10D, the compensation amount limit value I _th is set to a predetermined fixed value at time t20 when acceleration is started. In the present embodiment, the determination method of whether or not the steady state can be established and the content of control in the steady state are the same as those in the first embodiment.

That is, referring to FIGS. 10A and 10C, at time t22, the speed condition (the rotational speed ω _r is in the range of ω _r ^* ± Δω _rth ) and the axis error condition (estimated axis error Δθ _c is within the range of the steady state axis error determination value ± Δθ _th ), and the steady state is established at time t24 when the steady state determination time T _j has elapsed from time t22. When the limit value setting unit 228 determines that the steady state has been established, the compensation amount limit value I _th is changed to the q-axis current command average value I _qave ^* .

Also in the present embodiment, the steady state becomes unsatisfactory again according to any one of “mode 1: acceleration”, “mode 2: deceleration”, and “mode 3: large axial error”. However, in this embodiment, the same processing is performed in any aspect in that the compensation amount limit value I _th is returned to the above-mentioned “predetermined fixed value”. 10C and 10D, examples of waveforms when the axis error condition is not satisfied at time t26 (the estimated axis error Δθ _c exceeds the range of the steady state axis error determination value ± Δθ _th ). Show.

(Operation during deceleration)
Next, with reference to FIGS. 11A to 11D, an operation example during deceleration in the present embodiment will be described. FIGS. 11A to 11D are waveform diagrams showing examples of waveforms at various parts during deceleration of the motor control system of the present embodiment. In the example of FIG. 11A, since the rotational speed ω _r is higher than the rotational speed command ω _r ^{* at} time t30, the rotational speed ω _r decreases after time t30.

In FIG. 11 (d), the at the time the deceleration is started t30, the compensation amount limit value I _th is set to a predetermined fixed value as described above. The limit value setting unit 228 sequentially determines whether or not a steady state has been established, as in the above-described acceleration (FIGS. 10A to 10D). Referring to FIGS. 11A and 11C, at time t32, the speed condition (the rotational speed ω _r is in the range of ω _r ^* ± Δω _rth ) and the axis error condition (estimated axis error Δθ _c are to fall within the scope of the steady-state base error determination value ± [Delta] [theta] _th) is satisfied, the steady state is established at the time t34 to the steady state determination time T _j has elapsed from the time t32.

When the limit value setting unit 228 determines that the steady state has been established, the compensation amount limit value I _th is changed to the q-axis current command average value I _qave ^* . Even at the time of deceleration, the steady state is not established again by any one of “mode 1: acceleration”, “mode 2: deceleration”, and “mode 3: large axis error”. However, in any aspect, the same processing is performed in that the compensation amount limit value I _th is returned to the above-mentioned “predetermined fixed value”. 11C and 11D, examples of waveforms when the axis error condition is not satisfied at time t36 (the estimated axis error Δθ _c exceeds the range of the steady state axis error determination value ± Δθ _th ). Show.

As described above, in the present embodiment, like the first embodiment, since the PM motor 5 to set the compensation amount limit value I _th in accordance with whether or not a steady state, in the non-steady state PM The situation in which the behavior of the motor 5 becomes unstable can be effectively suppressed, power consumption can be reduced in the steady state, and the pulsation of the DC voltage V0 can be suppressed. Further, in the present embodiment, since the compensation amount limit value I _th in the unsteady state is set to a predetermined fixed value, there is an effect that resources for calculating the compensation amount limit value I _th can be reduced.

[Third Embodiment]
(Operation during acceleration)
Next, a third embodiment of the present invention will be described.
The configuration of the third embodiment is the same as that of the first embodiment (FIGS. 1 to 7), but the setting contents of the compensation amount limit value I _th are different. The contents will be described with reference to FIGS. FIGS. 12A to 12D are waveform diagrams showing examples of waveforms at various parts during acceleration of the motor control system of the present embodiment.

In the example of FIG. 12A, acceleration starts at time t40, and the rotational speed ω _r is lower than the rotational speed command ω _r ^*. Therefore, the rotational speed ω _r increases after time t40. At time t40 in FIG. 12D, the limit value setting unit 228 causes the q-axis current command average value I _qave ^{* to} be changed to the provisional compensation amount limit value I _th (that is, the compensation amount limit value I). set as the initial value of _th ).

At time t40 later, to be proportional to the increase rate of the rotation speed omega _r, compensation amount limit value I _th rate of increase is set. According to FIG. 12A, since the rotational speed ω _r increases linearly during the period from time t40 to t41, as shown in FIG. 12D, the compensation amount limit value I _th also linearly. It has increased. When the rotational speed ω _r and the rotational speed command ω _r ^* become substantially equal at time t41, the compensation amount limit value I _th also becomes a substantially constant value.

In the present embodiment, the method for determining whether or not a steady state can be established is the same as in the first embodiment. That is, in FIGS. 12A and 12C, at time t42, the speed condition (rotation speed ω _r is in the range of ω _r ^* ± Δω _rth ) and the axis error condition (estimated axis error Δθ _c are to fall within the scope of the steady-state base error determination value ± [Delta] [theta] _th) is satisfied, the steady state is established at the time t44 to the steady state determination time T _j has elapsed from the time t42.

However, the control contents in the steady state are also slightly different from those in the first embodiment. That is, in the present embodiment, the limit value setting unit 228 uses the compensation amount limit value I _th at the time point (t44) when the steady state is established as the start value and the value of the q-axis current command average value I _qave ^* as the end value. Then, a compensation amount limit value I _th that gradually decreases at a predetermined rate is set. In the example of FIG. 12D, the compensation amount limit value I _th reaches the q-axis current command average value I _qave ^* , which is the end value, at time t45.

Also in the present embodiment, the steady state becomes unsatisfactory again according to any one of “mode 1: acceleration”, “mode 2: deceleration”, and “mode 3: large axial error”. In this case, in this embodiment, the compensation amount limit value I _th is changed by the limit value setting unit 228 so that the same value as the value set in the first embodiment is set as the end value. However, in the first embodiment, the compensation amount limit value I _th is changed in a step shape, but in the present embodiment, there is a point that the current value gradually changes from the current value to the end value at a predetermined rate. Different. 12C and 12D, examples of waveforms when the axis error condition is not satisfied at time t46 (the estimated axis error Δθ _c exceeds the range of the steady state axis error determination value ± Δθ _th ). Show.

According to FIG. 12D, the compensation amount limit value I _th is compensated to a value just before the steady state is established (just before time t44 in the example of FIG. 12D), as in the first embodiment. The quantity limit value I _th is returned. However, it is instead of being returned to the step-like, within a period of time T46～t47, that compensation amount limit value I _th is gradually increased at a predetermined rate different from the first embodiment.

(Operation during deceleration)
Next, with reference to FIGS. 13A to 13D, an operation example during deceleration in the present embodiment will be described. FIGS. 13A to 13D are waveform diagrams showing examples of waveforms at various parts during deceleration of the motor control system of the present embodiment. In the example of FIG. 13A, the rotational speed ω _r is higher than the rotational speed command ω _r ^{* at} time t50, and therefore the rotational speed ω _r decreases after time t50.

In FIG. 13 (d), the at deceleration time is started t50, the compensation amount limit value I _th, as in the case of the first embodiment, q-axis current command average value I _qave at that time t50 ^* is compensated It is set to an amount limit value I _th. Also in the present embodiment, the compensation amount limit value I _th is kept constant until a steady state is established in preparation for an unexpected change in the load torque T _L.

The method for determining whether or not the steady state can be established during deceleration is the same as that during acceleration. That is, in FIGS. 13A and 13C, at time t52, the speed condition (the rotational speed ω _r is in the range of ω _r ^* ± Δω _rth ) and the axis error condition (estimated axis error Δθ _c are to fall within the scope of the steady-state base error determination value ± [Delta] [theta] _th) is satisfied, the steady state is established at the time t54 to the steady state determination time T _j has elapsed from the time t52.

The contents of control in the steady state are the same as those during acceleration. That is, compensation amount limit value I _th the starting value when the steady state has been established (t54), as the end value the value of the q-axis current command average value I _qave ^*, the limit value setting unit 228 gradually at a predetermined rate setting the compensation amount limit value I _th, as reduced to. In the example of FIG. 13D, the compensation amount limit value I _th reaches the q-axis current command average value I _qave ^* at time _t55 .

Even at the time of deceleration, the steady state is not established again by any one of “mode 1: acceleration”, “mode 2: deceleration”, and “mode 3: large axis error”. In this case, in this embodiment, the compensation amount limit value I _th is changed by the limit value setting unit 228 so that the same value as the value set in the first embodiment is set as the end value. However, in the first embodiment, the compensation amount limit value I _th is changed in a step shape, but in the present embodiment, there is a point that the current value gradually changes from the current value to the end value at a predetermined rate. Different. 13C and 13D, examples of waveforms when the axis error condition is not satisfied at time t56 (the estimated axis error Δθ _c exceeds the range of the steady state axis error determination value ± Δθ _th ). Show.

According to FIG. 13D, the compensation amount limit value I _th is compensated to a value just before the steady state is established (just before time t54 in the example of FIG. 13D), as in the first embodiment. The quantity limit value I _th is returned. However, it is instead of being returned to the step-like, within a period of time T56～t57, that compensation amount limit value I _th is gradually increased at a predetermined rate different from the first embodiment.

As described above, in the present embodiment, like the first embodiment, since the PM motor 5 to set the compensation amount limit value I _th in accordance with whether or not a steady state, in the non-steady state PM The situation in which the behavior of the motor 5 becomes unstable can be effectively suppressed, power consumption can be reduced in the steady state, and the pulsation of the DC voltage V0 can be suppressed. Further, in the present embodiment, since the gradually changed at a predetermined rate compensation amount limit value I _th upon transition steady state / non-steady-state, it is possible to prevent the control state is rapidly changed, control more stable The effect that it is made to play is also produced.

[Fourth Embodiment]
Next, a fourth embodiment of the present invention will be described.
The configuration of the fourth embodiment is the same as that of the first embodiment (FIGS. 1 to 7), but the setting contents of the compensation amount limit value I _th are different. The contents will be described with reference to FIGS. FIGS. 14A to 14D are waveform diagrams showing examples of waveforms at various parts during acceleration and deceleration of the motor control system of the present embodiment.

In the example of FIG. 14 (a), acceleration starts at time t60, and the rotational speed ω _r is lower than the rotational speed command ω _r ^{* at} that time, so the rotational speed ω _r increases after time t60. go. Deceleration starts at time t64, and the rotational speed ω _r is higher than the rotational speed command ω _r ^{* at} that time, so the rotational speed ω _r decreases after time t64. Further, at times t60 and t64 in FIG. 14D, the limit value setting unit 228 causes the provisional compensation amount limit value I _th (that is, the initial value of the compensation amount limit value I _{th) in} the same manner as in the first embodiment. ) Is set. That is, the product “K _v · I _qave ^* ” of the induced voltage change rate K _v and the q-axis current command average value I _qave ^* is set as the provisional compensation amount limit value I _th .

Further, at times t60 and t64, the calculation load torque T _LC shown in FIG. 14B is set to a value equal to the load torque T _L at that time. From time t60 to t62 and from time t64 to t66, when the rotational speed ω _r is increasing or decreasing (acceleration or deceleration), the limit value setting unit 228 calculates for each torque review period T _1. The load torque T _LC for use is updated to a value equal to the load torque T _L at that time. At this time, immediately before the calculation load torque T _LC is updated, the limit value setting unit 228 compares the load torque T _L with the calculation load torque T _LC . When the comparison result is “T _LC <T _L ”, it is desirable to increase the torque current, that is, the q-axis current command value I _q ^* from the current level, so that the compensation amount limit value I _th is increased from the current level. It is desirable.

Therefore, the limit value setting unit 228, if the comparison result is "T _LC <T _L", to the current compensation amount limit value I _th, the increment correction corresponding to the deviation (T _L -T _LC) performed, to the result as a new compensation amount limit value I _th. For example, as in the first embodiment, when the proportional constant and K _i, K _i and added value of (T _L -T _LC) with respect to the previous compensation amount limit value I _th, the new compensation The amount limit value I _th may be set.

On the other hand, if the comparative result is "T _LC ≧ T _L", in the present embodiment, the point of reducing the compensation amount limit value I _th in accordance with the deviation it is different from the first embodiment. For example, a value obtained by subtracting K _i ( _TLC − _TL ) from the previous compensation amount limit value I _th may be set as a new compensation amount limit value I _th . Further, the present embodiment is different from the first embodiment in that the distinction between the steady state and the unsteady state is not particularly performed.

In FIG. 14A, the rotational speed command ω _r ^* increases during the period from time t60 to t62, and the rotational speed command ω _r ^* decreases during the period from time t64 to t66. Accordingly, during these periods, the compensation amount limit value I _th is updated every torque review period T ₁ . On the other hand, the rotation speed command ω _r ^* is constant during the period from time t62 to t64. Accordingly, during this period, as shown in FIG. 14B, the calculation load torque T _LC is maintained at a constant value even though the load torque T _L varies, and FIG. 14C. As shown in FIG. 5, the compensation amount limit value I _th is also maintained at a constant value.

As described above, according to the present embodiment, the limit value setting unit 228 changes the compensation amount limit value I _th at each torque review period T ₁ when the rotational speed command ω _r ^* changes, and therefore, particularly the acceleration. Sometimes it becomes easy to cope with fluctuations in the load torque _TL . Further, during deceleration, an overcompensation state in which an excessive q-axis current command correction value I _qSIN ^* is output can be suppressed, and power consumption can be reduced.

[Modification]
The present invention is not limited to the above-described embodiments, and various modifications can be made. The above-described embodiments are illustrated for easy understanding of the present invention, and are not necessarily limited to those having all the configurations described. Further, a part of the configuration of an embodiment can be replaced with the configuration of another embodiment, and the configuration of another embodiment can be added to the configuration of an embodiment. Further, it is possible to delete a part of the configuration of each embodiment, or to add or replace another configuration. Examples of possible modifications to the above embodiment are as follows.

(1) In each of the above-described embodiments, an example in which a permanent magnet type synchronous motor is used as an electric motor has been described. Can be applied as well.

(2) The electric motor control system according to each of the above embodiments may be incorporated in an electric device such as an air conditioner or a refrigerator. Thereby, the control stability of these electric appliances and reduction of power consumption can be realized.

(3) In the above embodiments, the reproduced three-phase alternating currents I _uc , I _vc , and I _wc are obtained based on the current I 0 flowing through the current detector 7. The three-phase alternating currents I _u , I _v , and I _w may be directly measured using resistors or the like.

(4) When changing the compensation amount limit value I _th during acceleration (FIG. 8D) of the first embodiment and acceleration / deceleration (FIG. 14D) of the fourth embodiment, the previous compensation amount A correction value (K _i (T _L −T _LC )) was added to or subtracted from the limit value I _th . However, instead of subtraction may be changed compensation amount limit value I _th the increasing rate or decreasing rate in the previous compensation amount limit value I _th in the form of "multiply".

(5) In the above embodiments has been detected induced voltage through the induced voltage detection unit 24 may calculate the induced voltage by multiplying the rotation speed omega _r the EMF constant K _e.

(6) Since the hardware of the controller 2 in each of the above embodiments can be realized by a general computer, the program for realizing the block diagrams shown in FIGS. 1, 3, and 7 is stored in a storage medium, or You may distribute through a transmission line. The processing shown in FIGS. 1, 3, and 7 is a part or all of hardware using ASIC (Application Specific Integrated Circuit) or FPGA (field-programmable gate array). It may be realized by a typical process.

[Overview of composition and effect]
As described above, in the electric motor control device (2) of the first to fourth embodiments, the assumed rotation angle (θ _dc ) and actual value of the electric motor (5) that is rotationally driven by the inverter (3). An axis error estimator (15) for estimating an axis error (Δθ _c ) that is a difference from the rotation angle (θ _d ), and a torque command value (I _q ^* ) to suppress the axis error (Δθ _c ) ) And a torque controller (22) that outputs a torque command correction value (I _qSIN ^* ) that increases or decreases, and a drive signal to the inverter (3) based on the increased or decreased torque command value (I _q ^* ). A drive signal generator (14) that outputs (PWM signal), and the torque controller (22) sets a fluctuation range of the torque command correction value (I _qSIN ^* ) to a specified limit range (± correction value limiting unit that limits to fit I _th) and (227), said motor (5) Based on the state, and having a said limited range limit range setting unit to increase or decrease the _{(± I th) (228)} .

Thus, since the motor control device (2) sets the limit range (± I _th ) according to the state of the motor (5), the torque command correction value (I _qSIN ^* ) according to the state of the motor (5). Can be set. Therefore, when the state of the motor (5) fluctuates, the control can be stabilized by increasing the amplitude of the torque command correction value (I _qSIN ^* ), and the state of the motor (5) is stabilized. If it is, the power consumption can be reduced by limiting the fluctuation range of the torque command correction value (I _qSIN ^* ).

Furthermore, in the first to fourth embodiments, the electric motor (5) drives a load (6) whose load torque ( _TL ) fluctuates periodically, and the inverter (3) (5) applies a pulse-width modulated voltage, and the limit range setting unit (228) increases or decreases the limit range (± I _th ) according to the state of the load torque (T _L ). It is a thing to do. Thereby, the torque command correction value (I _qSIN ^* ) can be set according to the state of the load torque (T _L ).

Furthermore, the electric motor control device (2) of the first to fourth embodiments further includes a torque pulsation component output unit (21) for _obtaining a torque pulsation component (ΔT _mc ) based on the shaft error (Δθ _c ). The torque controller (22) includes a periodic component output unit (223, 224d, 224q) for obtaining a periodic component (ΔT _ds , ΔT _qs ) of the torque pulsation component (ΔT _mc ), and the periodic component ( And a compensator (33, 34, 225d, 225q) for supplying a compensation signal (output signal of 225d, 225q) for canceling (ΔT _ds , ΔT _qs ) to the correction value limiting unit (227). . Thus, the torque command correction value (I _qSIN ^* ) is set by the periodic component (ΔT _ds , ΔT _qs ) of the torque pulsation component (ΔT _mc ), and the variation range is set to the limit range (± I _th ). be able to.

Furthermore, in the electric motor control apparatus (2) of the first to fourth embodiments, the load (6) is a compression mechanism that compresses the refrigerant, and the electric motor (5) has a permanent magnet in its rotor. The rotation angle (θ _d , θ _dc ) indicates the magnetic pole position of the rotor, and the axis error estimation unit (15) is connected to the motor (5) from the inverter (3). ) Or the axis error (Δθ _c ) is estimated based on the current supplied to the inverter or the current flowing in the inverter (3). As a result, the axis error (Δθ _c ) can be estimated without measuring the current flowing through the electric motor (5).

Furthermore, in the third embodiment, the limit range setting unit (228) is characterized in that the limit range (± I _th ) is gradually changed at a predetermined rate. As a result, it is possible to prevent the control state from fluctuating rapidly, and the control can be further stabilized.

Furthermore, at the time of acceleration of the first embodiment and at the time of acceleration / deceleration of the fourth embodiment, the limit range setting unit (228) determines that the rotational speed (ω _r ) of the electric motor (5) is changing. The limiting range (± I _th ) is corrected, and when the rotational speed (ω _r ) is constant, the limiting range (± I _th ) is not corrected. Thereby, when the electric motor (5) is accelerating / decelerating, it becomes easy to cope with fluctuations in the load torque (T _L ) and the like.

Further, in the electric motor control device (2) of the first to third embodiments, the limit range setting unit (228) is configured such that the axial error (Δθ _c ) is equal to or longer than a predetermined time (T _j ). On the condition that it is within (± Δθ _th ), it is determined that the electric motor (5) is in the first state (steady state), while the axis error (Δθ _c ) is within the axis error range (± Δθ _th). ), The electric motor (5) is judged to be in the second state (unsteady state), and the limit range (±) is more than that in the case where the electric motor (5) is in the first state (steady state). I _th ) is widened.
Thereby, in the 2nd state (unsteady state), strong control can be performed with respect to an electric motor (5), and the situation where an electric motor (5) becomes unstable can be prevented beforehand. Further, when the electric motor (5) is in the first state (steady state), the fluctuation range of the torque command correction value (I _qSIN ^* ) can be narrowed, so that power consumption can be reduced.

Furthermore, in the electric motor control device (2) of the first to third embodiments, the limit range setting unit (228) further determines that the rotational speed (ω _r ) of the electric motor (5) is the predetermined time (T _j ) As described above, it is determined that the electric motor (5) is in the first state (steady state) on the condition that it is within a predetermined rotational speed range (ω _r ^* ± Δω _rth ), and the rotational speed ( Even when ω _r ) is out of the rotational speed range (ω _r ^* ± Δω _rth ), it is determined that the electric motor (5) is in the second state (unsteady state). As a result, the first and second states (steady state and unsteady state) can be determined based on both the axis error (Δθ _c ) and the rotational speed (ω _r ), thereby enabling more accurate control. Become.

  Moreover, in the electric equipment of the modification (2), the electric motor control device (2) according to claim 1 and the electric motor (5) are provided. As a result, when the state of the electric device fluctuates, the control can be stabilized, and when the state of the electric device is stable, the power consumption can be reduced.

1 Rotational speed command generator 2 Controller (motor controller)
3 Inverter 4 DC power supply 5 PM motor (motor)
6 Compression mechanism (load)
7 Current Detector 8 Current Reproducer 10 I _q0 ^* Generator 11 I _d ^* Generator 14 PWM Pulse Generator (Drive Signal Generator)
15 Δθ estimator (axis error estimator)
21 ΔT _m estimator (torque pulsation component output unit)
22 Torque controller 24 Induced voltage detector 223 Single phase-dq coordinate converter (periodic component output unit)
224d, 224q First-order lag filter (periodic component output unit)
225d, 225q integrator 226 dq-single phase inverse converter 227 limiter unit (correction value limiting unit)
228 Limit value setting part (Limit range setting part)

Claims (9)

An axis error estimation unit that estimates an axis error that is a difference between an assumed rotation angle and an actual rotation angle for an electric motor that is rotationally driven by an inverter;
A torque controller for outputting a torque command correction value for increasing or decreasing the torque command value in order to suppress the axis error;
A drive signal generator for outputting a drive signal to the inverter based on the increased or decreased torque command value, and the torque controller includes:
A correction value limiting unit that limits the fluctuation range of the torque command correction value so as to be within a specified limit range;
A motor control apparatus comprising: a limit range setting unit that increases or decreases the limit range based on a state of the motor. The electric motor drives a load whose load torque fluctuates periodically,
The inverter applies a pulse width modulated voltage to the electric motor,
The motor control device according to claim 1, wherein the limit range setting unit increases or decreases the limit range according to a state of the load torque. A torque pulsation component output unit for obtaining a torque pulsation component based on the axis error;
The torque controller is
A periodic component output unit for obtaining a periodic component of the torque pulsation component;
The motor control device according to claim 2, further comprising: a compensator that supplies a compensation signal that cancels the periodic component to the correction value limiting unit. The load is a compression mechanism that compresses the refrigerant,
The electric motor is a permanent magnet type synchronous motor having a permanent magnet in a rotor,
The rotation angle indicates a magnetic pole position of the rotor,
The motor control apparatus according to claim 3, wherein the shaft error estimation unit estimates the shaft error based on a current supplied from the inverter to the motor or a current flowing in the inverter. .   The motor control apparatus according to claim 1, wherein the limit range setting unit gradually changes the limit range at a predetermined rate. The limit range setting unit corrects the limit range when the rotation speed of the electric motor is changing, and does not correct the limit range when the rotation speed is constant. The electric motor control device described in 1. The limit range setting unit determines that the motor is in the first state on the condition that the axis error is within a predetermined axis error range for a predetermined time or more, while the axis error is within the axis error range. 2. The motor control device according to claim 1, wherein the motor control device determines that the motor is in a second state when the motor is out of the range, and makes the limit range wider than a case where the motor is in the first state. . The limit range setting unit further determines that the motor is in the first state on the condition that the rotation speed of the motor is within a predetermined rotation speed range for the predetermined time or more, and the rotation The electric motor control device according to claim 7, wherein it is determined that the electric motor is in the second state even when the speed is out of the rotational speed range. The motor control device according to claim 1;
An electric device comprising the electric motor.

Images