JP7247468B2 - motor controller - Google Patents

Description

The present invention relates to a motor control device.

In general, a motor control device that drives and controls a motor by position sensorless vector control generates a d-axis current command value and a q-axis current command value so that the rotation speed of the motor becomes a speed command value (target speed), A d-axis voltage command value is generated from the d-axis current command value, and a q-axis voltage command value is generated from the q-axis current command value. Furthermore, the motor control device converts the d-axis voltage command value and the q-axis voltage command value into three-phase voltage command values, and a PWM (Pulse Width Modulation) generator generates PWM based on the three-phase voltage command values. A signal is generated and output to, for example, an IPM (Intelligent Power Module). The IPM applies three-phase voltages (U-phase voltage Vu, V-phase voltage Vv, and W-phase voltage Vw) to the motor to drive and control the motor by performing switching control according to the input PWM signal.

Position sensorless vector control detects the position of the rotor of the motor in order to drive and control the motor. The voltage applied to the motor (voltage amplitude and frequency) for controlling the motor is adjusted based on the phase current (current amplitude and frequency) flowing through the motor. In recent years, for the purpose of cost reduction, a method of detecting a phase current without using a sensor for detecting the phase current has been used. One of these methods is current detection using a shunt resistor used to protect the bus current. From the current flowing through the shunt resistor (shunt current), the phase current of two phases is detected using the duty difference of the output voltages of the U, V, and W phases, and the remaining one phase is calculated from Kirchhoff's law. By calculating, the phase currents of the three phases are detected.

Furthermore, in recent years, from the viewpoint of energy saving, in order to reduce the switching loss of the inverter, control by two-phase modulation instead of three-phase modulation and overmodulation control for the purpose of expanding the maximum output range have been actively studied. there is However, two-phase modulation and overmodulation control increase the probability that the output voltage of each phase is full-duty, and the duty difference between phases disappears, making it difficult to detect phase currents using shunt currents. As a result, the period during which the shunt current can be accurately detected is reduced, making it difficult to continue motor drive control. Therefore, a control that can continue the drive control of the motor even when the shunt current cannot be detected is being studied.

JP 2009-124782 A JP 2009-261066 A JP 2015-12770 A

As a basic method of estimating the phase current of a phase that cannot be detected when the shunt current cannot be detected, as shown in FIG. There is a method of estimating the phase current based on motor parameters (inductance value, resistance value, induced voltage constant), motor rotation speed (rotation speed), and rotor rotation phase.

The feature of this method is that the phase current is reproduced by estimating from the phase voltage, motor parameters, rotation speed, and rotation phase, so position sensorless vector control can be performed without changing the control configuration of the motor control device. . However, phase current estimation requires accuracy similar to that of the prior art.

In the prior art described in Patent Document 1, as shown in FIG. 1, the stator of the motor has inductances Lu, Lv, and Lw of each phase of U, V, and W phases and resistances Ru, Rv. , Rw. Then, the currents Iu, Iv, and Iw of each phase are calculated from the applied voltages (phase voltages) Vu, Vv, and Vw, the induced voltages Eu, Ev, and Ew, and the neutral point potential Vn (potential at the connection point of each phase), It is said that it can be realized by multiplying the ratio of estimated two-phase currents that cannot be detected by the one-phase current that can be detected. Estimated currents Iue, Ive, and Iwe of each phase in this case are as shown in the following (Equation 1) to (Equation 3). The following (Equation 1) to (Equation 3) are current equations in an equivalent circuit of a general motor. Note that “s” in the following (formula 1) to (formula 3) is the Laplace operator.

Iue=(Vu−Vn−Eu)/(s·Lu+Ru) (Formula 1)
Ive=(Vv−Vn−Ev)/(s·Lv+Rv) (Formula 2)
Iwe=(Vw−Vn−Ew)/(s·Lw+Rw) (Formula 3)

The above-mentioned "ratio (ratio) by the estimated current of the two phases that cannot be detected" means that the estimated current of the two phases that cannot be detected is calculated, and the estimated current of each phase of the estimated current of the two phases is calculated as the estimated current of the two phases. It is divided by the sum. By multiplying the detected one-phase current by this division result, the three-phase current is reproduced. For example, when only the U-phase current Iu can be detected, the estimated currents of the other two phases can be obtained from the following (Equation 4) to (Equation 5).

Iv=−Iu・Ive/(Ive+Iwe) (Formula 4)
Iw=−Iu·Iwe/(Ive+Iwe) (Formula 5)

Thus, in the method described in Patent Document 1, the detected current is obtained by multiplying the detected one-phase current by the ratio of the undetected two-phase estimated current. At this time, whether or not the estimated current value is the same as the undetected current (real phase current) cannot be determined based on the ratio of the undetected estimated currents of the two phases. In addition, the ratio of the two-phase estimated currents that could not be detected has the sum of the two-phase estimated currents in the denominator. There is a problem that the two-phase current values that cannot be detected cannot be calculated correctly because division by the value results in infinity.

In addition, the current equations (Equation 1) to (Equation 3) in the equivalent circuit of a general motor include the Laplace operator s. , the term of the Laplace operator s can be ignored. However, in an actual operating environment, there are timings (intervals) in which the current value is not stable due to various factors such as the influence of disturbances and fluctuations in the load torque of the motor. cannot be ignored. Therefore, in normal control, it is necessary to perform calculations (Laplace transforms) including the Laplace operator s in the above (Equation 1) to (Equation 3), which poses a problem that the amount of calculation may become enormous.

There is also a method of estimating current on two axes (dq axes) by open loop control, as in the prior art described in Patent Document 2 mentioned above. However, unlike feedback control, which acquires and controls motor operating information (phase current, phase voltage, rotation speed, etc.) in real time, open loop control cannot grasp the actual phase current, so it is There is a problem that there is no guarantee that optimum motor control can be performed.

Further, as in the prior art described in Patent Document 3, the solution of the calculation formula including the Laplace operator s is obtained in advance, and the calculated solution and the motor performance under the actual operating environment are calculated in advance. There is also a method of storing in a lookup table in association with parameters indicating actual operating conditions. In this method, a lookup table is referenced based on parameters indicating the actual operating conditions of the motor to obtain a solution of a calculation formula including the Laplace operator s. However, there is no guarantee that the solutions obtained in advance can be handled as correct values under all actual operating environments, and there is no guarantee that optimum motor control can be performed under actual operating environments. In addition, there is also the problem that a storage device is required to store the lookup table, and the capacity of the storage device may become enormous depending on the actual operating conditions, such as when a plurality of conditions are included.

Further, when the estimated current is calculated from the estimated induced voltage, highly accurate estimation cannot be performed unless the induced voltage constant, rotational speed, and phase, which are motor parameters, are correct. Furthermore, the induced voltage constant is affected by individual variations in motors, temperature dependency of magnets in motors, and deterioration over time, which affects the accuracy of the estimated current.

As described above, according to the above-described conventional technology, the problem of the accuracy of the estimated current and the problem of the huge amount of calculation required to increase the accuracy of the estimated current can be solved efficiently and accurately. Therefore, there is a possibility that optimal and stable motor control cannot be continuously performed under the actual operating environment.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a motor control device capable of efficiently and accurately obtaining a current used to drive a motor.

In order to solve the above-described problems, the motor control device according to the embodiment of the present invention includes, for example, a drive section, a detection section, and a drive voltage generation section. The driving unit drives the motor by supplying a driving voltage generated based on the difference between the target speed and the current speed of the motor to the motor. The detector detects a phase current of each phase in the UVW coordinate system that flows through the motor. The driving voltage generator generates a driving voltage for each phase in the UVW coordinate system from the driving voltage for each phase in the dq coordinate system. The detector includes a current estimator that estimates the phase current of each phase in the UVW coordinate system that flows through the motor. The current estimator estimates the phase current of each phase in the UVW coordinate system that flows through the motor. The current estimator calculates the induced voltage of each phase of the motor based on the voltage command value of each phase in the dq coordinate system, and calculates the phase current of each phase in the UVW coordinate system flowing through the motor using the calculated induced voltage. to estimate.

According to one embodiment of the present invention, it is possible to efficiently and accurately obtain the current used to drive the motor.

FIG. 1 is a diagram showing an example of the configuration of an equivalent circuit of a motor. FIG. 2 is a diagram conceptually showing the estimated current without using the Laplace transform. FIG. 3 is a diagram for explaining the dq-axis and the αβ-axis. FIG. 4 is a diagram showing an example of the configuration of the motor control device according to the first embodiment. FIG. 5 is a diagram showing an example of the configuration of the 3φ current calculator according to the first embodiment. FIG. 6 is a diagram for explaining current detection determination, d-axis current, and q-axis current at processing timings. FIG. 7 is a basic vector diagram in the motor model. FIG. 8 is a diagram for explaining calculation of the induced voltage on the dq axis. FIG. 9 is a diagram showing an example of the configuration of a motor control device according to the second embodiment. FIG. 10 is a diagram showing an example of the configuration of a 3φ current calculator according to the second embodiment. FIG. 11 is a diagram illustrating an example of the configuration of an estimated current calculator according to the second embodiment;

An embodiment of a motor control device of technology disclosed in the present application will be described in detail below with reference to the drawings. Note that the disclosed technology is not limited by the following embodiments. In the following embodiments, the motor control device will be described as a control device for a motor whose load is a propeller fan, compressor, or the like used in an air conditioner or the like. Applicable to general control. Each of the following embodiments can be implemented in combination as appropriate within a non-contradictory range.

Also, in the following embodiments, the configuration and processing according to the disclosed technology will be mainly described, and descriptions of other configurations and processing will be simplified or omitted as appropriate. Moreover, in the following embodiments, the same reference numerals are assigned to the same configurations and processes, and the descriptions of the already-outed configurations and processes are omitted.

First, the outline of the first embodiment will be explained. Embodiment 1 is a motor control device that can calculate a highly accurate estimated current from a detected current of one phase without depending on the actual operating environment of the motor without greatly changing the control configuration of the basic vector control. In the first embodiment, the estimated current expressed in the voltage dimension (current in units of voltage [V]) is obtained from the motor output voltage command value and the estimated induced voltage of the motor, and the detected current of one phase that can be detected is Using the current amplitude of , the drive control of the motor is performed with an estimated current (current whose unit is current [A]) converted from the voltage dimension to the current dimension.

In addition, in Embodiment 1, a one-shunt current detection method is used. However, in the 2CT current detection method in which two phase currents are detected by a 2CT (Current Transformer) and the remaining phase currents are calculated from the relational expression Iu + Iv + Iw = 0 of Kirchhoff's law, one of the two phase sensors fails, Even if only one-phase current can be detected, the estimated current can be calculated in the same manner as the one-shunt current detection method.

Embodiment 1 reproduces (estimates) three-phase currents in order to perform motor drive control in the same manner as two-phase current detection without significantly changing the control configuration of the basic vector control of the motor control device. ), or reproduce (estimate) the dq-axis currents after converting the three-phase currents. FIG. 1 is a diagram showing an example of the configuration of an equivalent circuit of a motor. As shown in FIG. 1, the stator of the motor is replaced with an equivalent circuit of inductances Lu, Lv, Lw and resistors Ru, Rv, Rw in each phase (U, V, W phases), and the applied voltage ( Phase voltages) Vu, Vv, Vw, induced voltages Eu, Ev, Ew, and neutral point potential Vn are used to calculate currents Iu, Iv, and Iw of each phase. Calculations of (Equation 6) to (Equation 8) are performed. Note that “s” in the following (Equation 6) to (Equation 8) is a Laplace operator.

Iue=(Vu−Vn−Eu)/(s·Lu+Ru) (Formula 6)
Ive=(Vv−Vn−Ev)/(s·Lv+Rv) (Formula 7)
Iwe=(Vw−Vn−Ew)/(s·Lw+Rw) (Formula 8)

The applied voltage, induced voltage, neutral point potential, inductance value, and resistance value of each phase included in the above (Equation 6) to (Equation 8) are parameters that constantly fluctuate during operation. In particular, since inductance and resistance values have current characteristics and temperature characteristics, in order to ensure the accuracy of current estimation, real-time processing such as correction for changes due to characteristics or online identification is required as necessary. Become.

Formulas for calculating the induced voltages Eu, Ev, and Ew when it is assumed that the induced voltages of the U-phase, V-phase, and W-phase appear in a sinusoidal form are shown in (Equation 9) to (Equation 11) below.

Eu=Ke·sin(θ)·ω (Formula 9)
Ev=Ke·sin(θ−2×π/3)·ω (Formula 10)
Ew=Ke·sin(θ+2×π/3)·ω (Formula 11)

"Ke" in the above (Equation 9) to (Equation 11) indicates the induced voltage constant [V/rpm], "θ" indicates the rotor rotation angle [rad], and "ω" indicates the motor rotation speed ( rotation speed) [rpm]. If the units of the induced voltage constant Ke, the rotor rotation angle θ, and the motor rotation speed (rotational speed) ω are represented by [V], then what kind of unit combinations can be used? It's okay. Since the induced voltage constant Ke is a motor constant, it is a known parameter. The rotation angle θ may be obtained from a sensor such as a Hall element, or may be calculated from the estimated rotation speed (rotation speed) ω. The rotational speed (rotational speed) ω may be obtained from a sensor such as a resolver, or may be calculated from the difference between the actual rotor position and the rotor position held by the motor control device.

Here, the neutral point potential Vn can be calculated from the following (Equation 12).

Vn=((Vu+Vv+Vw)-(Eu+Ev+Ew))/3 (Formula 12)

It should be noted that the applied voltages (phase voltages) Vu, Vv, and Vw in the above (Equation 12) are obtained by a method of providing voltage sensors at the motor terminals of the U-phase, V-phase, and W-phase, or by a motor control device. Output voltage command values Vu ^* , Vv ^* , Vw ^* held inside the microcomputer may be used.

Then, for example, when only the U-phase current Iu can be detected, the estimated currents of the other two phases can be obtained from the following (Equation 13) to (Equation 14). If only the V-phase current Iv can be detected, replace Iv on the left side of Equation 13 below with Iu, and replace Iu on the right side of Equations 13 and 14 with Iv, and replace Ive with Iue. , the estimated currents of the other two phases that could not be detected can be obtained. Further, when only the current Iw of the W phase can be detected, Iw on the left side of the following (Equation 14) is replaced with Iu, and Iu on the right side of (Equation 13) and (Equation 14) is replaced with Iw, and Iwe is replaced with Iue. By replacing, the estimated currents of the other two phases that could not be detected can be obtained.

Iv=−Iu·Ive/(Ive+Iwe) (Formula 13)
Iw=−Iu·Iwe/(Ive+Iwe) (Formula 14)

When the undetectable two-phase current is obtained from the ratio of the detected one-phase current and the undetectable two-phase current as in the above (Equation 13) to (Equation 14), (Equation 6) (Equation 8), calculation including the Laplacian operator s must be performed, requiring an enormous amount of calculation. Therefore, in the first embodiment, in the process of calculating the estimated current, a method that eliminates the influence of motor parameters that change due to motor operation (such as the current characteristics of the inductance value and the temperature characteristics of the resistance) and does not include the Laplace operator s to calculate the estimated current.

(Calculation of estimated current not including Laplace operator s)
Next, an estimated current calculation that does not include the Laplace operator will be described. FIG. 2 is a diagram conceptually showing the estimated current without using the Laplace transform. FIG. 2(a) shows the applied voltage (phase voltage), and FIG. 2(b) shows the induced voltage. As shown in FIG. 2(a), the applied voltages Vu, Vv, and Vw of the U-phase, V-phase, and W-phase are sinusoidal waves having a certain magnitude of amplitude, and each has a phase difference of 120°. , the amplitude value changes according to the phase.

Therefore, the applied voltage (phase voltage), the neutral point potential, and the induced voltage can be regarded as components (phase components) that change according to the phase. On the other hand, unlike the applied voltage (phase voltage), the inductance value of each phase of U phase, V phase, and W phase, and the resistance value of each phase are not components whose value changes according to the phase, but It is an immutable constant value.

The Laplace operator s is an operator with temporal properties, but since the Laplace operator s changes the magnitude of the object to be acted upon, the amplitude can be seen as an ingredient. Therefore, by replacing the amplitude component including the Laplace operator s with the amplitude component not including the Laplace operator s, an estimated current not including the Laplace operator s can be realized. The factor part including the Laplace operator s in the above (Equations 6) to (Equation 8) is regarded as an amplitude component, and the amplitude component is replaced with a current amplitude correction coefficient Ka that does not include the Laplace operator s Three-phase estimated current Iue, Ive, and Iwe are shown in (Formula 15) to (Formula 17) below.

Iue=Ka·(Vu−Vn−Eu) (Equation 15)
Ive=Ka·(Vv−Vn−Ev) (Formula 16)
Iwe=Ka·(Vw−Vn−Ew) (Equation 17)

The current amplitude correction coefficient Ka included in the above (Equation 15) to (Equation 17) is an estimated current in voltage dimension based on the output voltage (Vu, Vv, Vw) (current in units of voltage [V]) is the amplitude information of the estimated current obtained by taking the ratio of the detected actual current amplitude to the amplitude of , and has the same value for each phase. Here, the output voltages (Vu, Vv, Vw) are applied voltages (phase voltages) applied to the motor.

(Estimated current in voltage dimension (current with unit as voltage [V])
Next, the estimated current in the voltage dimension will be described. The following equations (18) to (20) indicate the current having information on the phase component of the current, with the unit being voltage [V]. This is the phase component ((applied voltage)--neutral point potential--induced voltage) of the above equations (6) to (8). Therefore, since all the elements are voltages, it is defined as a current whose unit is voltage [V].

Iue_v=Vu-Vn-Eu (Formula 18)
Ive_v=Vv-Vn-Ev (Formula 19)
Iwe_v=Vw-Vn-Ew (Formula 20)

Estimated voltage-dimensional currents (currents in units of voltage [V]) represented by (Equation 18) to (Equation 20) are converted by 3Φ/2Φ and converted on the αβ axis (Equation 21) to ( 22).

Iαv=√(2/3)·(Iue_v−Ive_v/2−Iwe_v/2)
... (Formula 21)
Iβv=√(2/3)・(√(3)・Ive_v/2−√(3)・Iwe_v/2)
... (Formula 22)

The magnitude Kva of the voltage-dimensional current amplitude (current amplitude whose unit is voltage [V]) at this time can be obtained as shown in the following (Equation 23).

Kva=√(Iαv̂2+Iβv̂2) (Formula 23)

That is, Kva is the square root of the sum of Iαv squared and Iβv squared.

The current amplitude on the two axes (dq-axis current amplitude) Kdq when the current can be detected can be obtained as shown in (Equation 24), similarly to (Equation 23).

Kdq=√(Id̂2+Iq̂2) (Formula 24)

That is, Kdq is the square root of the sum of Id squared and Iq squared.

An amplitude component (current amplitude correction coefficient Ka) that does not include the Laplace operator s is calculated from the above (Equation 23) to (Equation 24).

The magnitude of the voltage-dimensional current amplitude (current amplitude with the unit as voltage [V]) is the magnitude of the current amplitude on the αβ axis obtained from (Equation 21) to (Equation 22) to (Equation 23), The magnitude of the current amplitude on the two axes when the current can be detected is on the dq axis. The relationship between these two axes is the difference between a stationary coordinate system and a rotating coordinate system. FIG. 3 is a diagram for explaining the dq-axis and the αβ-axis. As shown in FIG. 3, the dq-axis and the αβ-axis exist on the same plane, and the dq-axis exists at a position rotated by θdq [rad] with respect to the αβ-axis. (Equation 23) to (Equation 24) are the magnitude (scalar quantity) of the current vector, and do not have phase information or axis information such as the dq axis, the αβ axis, and the UVW axis. can be calculated regardless of the information in The current amplitude correction coefficient Ka can be obtained as shown in (Equation 25) below.

Ka=Kdq/Kva (Formula 25)

In this way, as shown in the above (formula 15) to (formula 17), the phase components shown in the above (formula 18) to (formula 20) are estimated by multiplying the current amplitude correction coefficient Ka, which is the amplitude component. Current can be calculated. However, this phase component is a value on the three axes (UVW phase), the magnitude of the current amplitude on the two axes (dq-axis current amplitude) Kdq, and the voltage-dimensional current amplitude (unit: voltage [V]). The axis information is different from the current amplitude correction coefficient Ka calculated with the magnitude Kva of the current amplitude (current amplitude).

Therefore, the magnitude of the current amplitude Kdq may be calculated by converting the above (Equation 23) to (Equation 24) into values on three axes before calculating the current amplitude correction coefficient Ka. is the ratio of scalar quantities that do not contain axis information. That is, Kdq is the magnitude of current on two axes (scalar on two axes), and Kva is also the magnitude of current on two axes (scalar on two axes). Therefore, since the current amplitude correction coefficient Ka is a coefficient obtained by taking a ratio on two axes as in (Equation 25), it is only necessary to multiply the aforementioned phase component. In addition, Ka has a unit of [A / V], but the phase components shown in (Equation 15) to (Equation 17) (or (Equation 18) to (Equation 20)) are voltage dimensions, and the unit is [ V], the unit of the multiplication result is [A], which is the current dimension.

(Replacement principle for current estimation that does not include Laplace operator s)
Next, the principle of replacing the current estimation including the Laplace operator s with the current estimation not including the Laplace operator s will be described. A motor current (phase current) flows according to the applied voltage for driving the motor. The detected three-phase motor currents (phase currents) are defined as detected currents (Iud, Ivd, Iwd). If this detected current can approximate the estimated current shown in the above (Equation 6) to (Equation 8), the above (Equation 6) to (Equation 8) can be expressed as the following (Equation 26) to (Equation 28). Become.

Iud≈Iue=(Vu−Vn−Eu)/(s·Lu+Ru) (Equation 26)
Ivd≈Ive=(Vv−Vn−Ev)/(s·Lv+Rv) (Equation 27)
Iwd≈Iwe=(Vw−Vn−Ew)/(s·Lw+Rw) (Equation 28)

As shown in (Equation 26) to (Equation 28) above, the motor current (phase current) is based on the phase and magnitude of the applied voltage, and therefore includes phase information and amplitude information. Components of the phase information are the aforementioned applied voltages (phase voltages) Vu, Vv, Vw, neutral point potential Vn, and induced voltages Eu, Ev, Ew. The components of the amplitude information are the inductances Lu, Lv, Lw of each phase, the Laplace operator s, and the resistance values Ru, Rv, Rw of each phase.

In order to process the 3-phase detected currents in a processor such as a microcomputer, they are converted into dq-axis currents on two axes according to the following equations (29) to (30).

Id=√(2/3)·{(Iud−Ivd/2−Iwd/2)·cos θ
+√(3)/2·(Ivd−Iwd)·sin θ} (Formula 29)
Iq=√(2/3)·{−(Iud−Ivd/2−Iwd/2)·sin θ
+√(3)·2·(Ivd−Iwd)·cos θ} (Formula 30)

The above (Equation 29) to (Equation 30) include the U-phase detection current Iud, the V-phase detection current Ivd, and the W-phase detection current Iwd, respectively, and the only other parameter is the phase θ. This phase θ is positional information of the rotor of the motor, and is an electrical angle phase θe on the dq axis output by a position estimator 30 (see FIG. 4) described later. If this electrical angle phase θe is accurate, it is considered that there is no conversion error due to the above (Equation 29) to (Equation 30). That is, it can be said that the current amplitude on the dq axis calculated by the above (Equation 29) to (Equation 30) has information on the current amplitude on the three-phase axis (UVW axis).

The amplitude component (current amplitude correction coefficient Ka) that does not include the Laplace operator s, obtained by (Equation 23) to (Equation 25), includes amplitude information of the detected current. Also, the amplitude information included in the detected current is equivalent to the information by the component including the Laplace operator s, as shown in (Eq.26) to (Eq.28). From these facts, the current amplitude correction coefficient Ka can be regarded as equivalent to information by a component including the Laplace operator s, as shown in the following (Equation 31). The following (Equation 31) expresses the current amplitude correction coefficient Ka using the U-phase inductance Lu and resistance Ru. , can similarly represent the current amplitude correction factor Ka.

Ka=Kdq/Kva≈1/(s·Lu+Ru) (Formula 31)

From the above, the amplitude information of the detected current can be replaced with a coefficient that does not include the Laplacian operator s.

<Regarding the configuration of the motor control device according to the first embodiment>
Embodiment 1 based on the theoretical background explained above will be explained. FIG. 4 is a diagram showing an example of the configuration of the motor control device according to the first embodiment.

A motor control device 100A of Embodiment 1 has a microcomputer 10A, an IPM (Intelligent Power Module) 23, and a one-shunt current detector 32. FIG. A motor 1 is connected to the motor control device 100A.

Also, the microcomputer 10A includes a subtractor 11, a speed controller 12, an excitation current controller 13, a subtractor 14, a subtractor 15, a d-axis current controller 16, a q-axis current controller 17, and a decoupling controller 18. , subtractor 19, adder 20, dq/3φ converter 21, PWM (Pulse Width Modulation) generator 22, 3φ current calculator 24A, 3φ/dq converter 25, axis error processor 26, PLL (Phase Locked Loop) controller 29 , position estimator 30 and 1/Pn processor 31 .

The subtractor 11 calculates the current ^rotational speed of the motor 1 (mechanical angle A speed deviation (mechanical angular speed deviation) Δω obtained by subtracting the estimated speed) ω is output to the speed controller 12 .

Speed controller 12 generates a q-axis current command value Iq ^* that makes speed deviation Δω output from subtractor 11 smaller, and outputs it to excitation current controller 13 and subtractor 15 . The excitation current controller 13 generates a d-axis current command value Id ^* from the q-axis current command value Iq ^* output from the speed controller 12 and outputs the same to the subtractor 14 . The d-axis and q-axis represent the coordinate axes of a two-phase rotating coordinate system (current vector coordinates), and Id, Iq, and Vd and Vq, which will be described later, represent current and voltage on these coordinate axes. The two-phase rotating coordinate system is also called a dq coordinate system.

A subtractor 14 subtracts the d-axis current Id output from the 3φ/dq converter 25 from the d-axis current command value Id ^* output from the excitation current controller 13 to generate a d-axis current deviation ΔId. Output to the shaft current controller 16 . Subtractor 15 subtracts q-axis current Iq output from 3φ/dq converter 25 from q-axis current command value Iq ^* output from speed controller 12 to generate q-axis current deviation ΔIq. Output to the current controller 17 .

A d-axis current controller 16 generates a d-axis voltage command value Vd ^** from the d-axis current deviation ΔId output from the subtractor 14 . A q-axis current controller 17 generates a q-axis voltage command value Vq ^** from the q-axis current deviation ΔIq output from the subtractor 15 .

A non-interfering controller 18 cancels the interference of the d-axis and the q-axis and generates a non-interfering correction value for controlling each independently. Specifically, the decoupling controller 18 calculates the d-axis voltage command value Vd ^* from the d-axis current Id output from the 3φ/dq converter 25 and the estimated electrical angle speed ωe output from the PLL controller 29. A d-axis decoupling correction value Vda for decoupling ^* is generated and output to the subtractor 19 . In addition, the non-interfering controller 18 converts the q-axis voltage command value Vq ^** into a A q-axis non-interfering correction value Vqa for interfering is generated and output to the adder 20 .

A subtractor 19 subtracts the d-axis decoupling correction value Vda output from the decoupling controller 18 from the d-axis voltage command value Vd ^** output from the d-axis current controller 16 to obtain a d-axis voltage A d-axis voltage command value Vd ^* is generated by decoupling the command value Vd ^** , and is output to the dq/3φ converter 21 and the axis error arithmetic processor . The adder 20 adds the q-axis decoupling correction value Vqa output from the decoupling controller 18 to the q-axis voltage command value Vq ^** output from the q-axis current controller 17 to obtain the q-axis voltage A q-axis voltage command value Vq ^* is generated by decoupling the command value Vq ^** , and is output to the dq/3φ converter 21 and the axis error arithmetic processor .

The dq/3φ converter 21 uses the electrical angle phase (dq-axis phase) θe, which is the current rotor position output from the position estimator 30, to generate a non-interfering two-phase d-axis voltage command value Vd ^* and q-axis voltage command value Vq ^* are converted into U-phase output voltage command value Vu ^* , V-phase output voltage command value Vv ^* , and W-phase output voltage command value Vw ^* , which are three-phase voltage command values. Then, the dq/3φ converter 21 outputs the U-phase output voltage command value Vu ^* , the V-phase output voltage command value Vv ^* , and the W-phase output voltage command value Vw ^* to the PWM generator 22 and the 3φ current calculator 24A. . U-phase output voltage command value Vu ^* , V-phase output voltage command value Vv ^* , W-phase output voltage command value Vw ^* , and U-phase detection current Iu, V-phase detection current Iv, and W-phase detection current, which will be described later. Iw is the three-phase fixed coordinate system voltage and current.

The PWM generator 22 generates 6-phase PWM signals from the U-phase output voltage command value Vu ^* , the V-phase output voltage command value Vv ^* , the W-phase output voltage command value Vw ^* , and the PWM carrier signal, and supplies them to the IPM 23. Output. PWM generator 22 is an example of a signal generator. Note that the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* may be used as the voltage command values, and the dq/3φ converter 21 may be included in the signal generator.

Based on the 6-phase PWM signal output from the PWM generator 22, the IPM 23 generates AC voltages to be applied to the U-phase, V-phase, and W-phase of the motor 1 from the externally supplied DC voltage Vdc. Then, the AC voltages are applied to the U-phase, V-phase, and W-phase of the motor 1, respectively. The IPM 23 is an example of a drive unit that supplies the motor 1 with a drive voltage generated based on the difference between the target speed and the current speed of the motor 1 to drive the motor 1 . The IPM 23 may be an IC (Integral Circuit) in which transistors and diodes are integrated, for example, but may also be configured by arranging each component on a circuit board, for example.

The 1-shunt current detector 32 detects a bus current flowing through a shunt resistor (not shown) by a 1-shunt current detection method. Then, the 1-shunt current detector 32 receives information on the detected bus current and switching information on the 6-phase PWM signals (Up, Vp, Wp, Un, Vn, Wn) output from the PWM generator 22, Output to the 3φ current calculator 24A. Note that the 1-shunt current detector 32 uses two current transformers (CTs) of the U-phase, V-phase, and W-phase of the motor 1 to detect two-phase currents. may be substituted.

The 3φ current calculator 24A calculates U Phase detection current Iu, V-phase detection current Iv, and W-phase detection current Iw are calculated. Then, the 3φ current calculator 24 A outputs the calculated U-phase detected current Iu, V-phase detected current Iv, and W-phase detected current Iw of the motor 1 to the 3φ/dq converter 25 .

Alternatively, when the 3φ current calculator 24A detects the bus line current by the 2CT current detection method, it calculates the other one-phase current of the detected two-phase current from the relational expression Iu+Iv+Iw=0 of Kirchhoff's law.

Details of the configuration and processing of the 3φ current calculator 24A will be described later. The 3φ current calculator 24A is a part of the microcomputer and is an example of a detection section that detects the current flowing through the motor 1 .

The 3φ/dq converter 25 uses the electrical angle phase θe output from the position estimator 30 to convert the U-phase detection current Iu, the V-phase detection current Iv, and the three-phase detection current Iu output from the 3φ current calculator 24 . The W-phase detected current Iw is converted into two-phase d-axis current Id and q-axis current Iq. Then, the 3φ/dq converter 25 passes the d-axis current Id to the subtractor 14, the decoupling controller 18, the 3φ current calculator 24A, the axis error arithmetic processor 26, and the q-axis current Iq to the subtractor 15, the Output to the interfering controller 18, the 3φ current calculator 24A, and the axis error processor 26, respectively.

The axis error processor 26 calculates the d-axis voltage command value Vd ^* output from the subtractor 19, the q-axis voltage command value Vq ^* output from the adder 20, and the d An axis error variation Δθ is calculated from the axis current Id and the q-axis current Iq and output to the PLL controller 29 . Here, the axis error is the deviation between the actual dq-axis and the control dq-axis (γδ-axis).

The PLL controller 29 calculates an estimated electrical angle speed ωe, which is the estimated current angular speed of the rotation of the motor 1, from the axis error variation Δθ output from the axis error processor 26, and the non-interfering controller 18 , 3φ current calculator 24A, position estimator 30, and 1/Pn processor 31, respectively.

The position estimator 30 calculates an electrical angle phase (dq axis phase) θe, which is an estimated position obtained by estimating the rotor position of the motor 1, from the electrical angle estimated speed ωe output from the PLL controller 29 . The position estimator 30 then outputs the electrical angle phase θe to the dq/3φ converter 21 and the 3φ/dq converter 25, respectively. The position estimator 30 is an example of a position estimator that calculates the estimated rotor position of the motor 1 .

The 1/Pn processor 31 divides the estimated electrical angle speed ωe output from the PLL controller 29 by the pole logarithm Pn of the motor 1 to calculate the current rotation speed (mechanical angular speed) ω of the motor 1, and the subtractor 11.

<Regarding the configuration of the 3φ current calculator of the first embodiment>
FIG. 5 is a diagram showing an example of the configuration of the 3φ current calculator according to the first embodiment. As shown in FIG. 5, the 3φ current calculator 24A has a detected current phase/detected pattern determination processor 24Aa, a detected current processor 24Ab, an estimated current calculator 24Ac as a current estimator, and a current reproduction processor 24Ad.

The detected current phase/detected pattern determination processing unit 24Aa determines which phase the detected current output from the one-shunt current detector 32 belongs to based on the switching information of the six-phase PWM signals output from the PWM generator 22. Based on the determination result, a detection current phase, which is information indicating the phase of the detection current, and a detection pattern, which is information indicating the number of phases of the detection current, are generated. Then, the detected current phase/detected pattern determination processing section 24Aa outputs the detected current phase to the detected current processing section 24Ab, and outputs the detected pattern to the current reproduction processing section 24Ad.

Based on the detected current phase output from the detected current phase/detected pattern determination processing unit 24Aa, the detected current processing unit 24Ab determines the detected current detected by the one-shunt current detector 32 using a shunt resistor (not shown). is the U-phase detection current Iud, the V-phase detection current Ivd, or the W-phase detection current Iwd. Then, the detected current processing unit 24Ab outputs the specified U-phase detected current Iud, V-phase detected current Ivd, and W-phase detected current Iwd to the current reproduction processing unit 24Ad. When none of the currents is detected by the one-shunt current detector 32 using a shunt resistor (not shown), the detected current processing unit 24Ab detects the U-phase detected current Iud and the V-phase detected current Ivd. , W-phase detection currents Iwd are not output to the current reproduction processing unit 24Ad.

The estimated current calculator 24Ac outputs the U-phase output voltage command value Vu ^* , the V-phase output voltage command value Vv ^* , the W-phase output voltage command value Vw ^* output from the dq/3φ converter 21, and the 3φ/dq converter 25. The d-axis current Id output from the q-axis current Iq, the electrical angle estimated speed ωe output from the PLL controller 29, the electrical angle phase θe output from the position estimator 30, the externally given such as a storage device Based on the induced voltage constant Ke, which is a motor parameter, the above (Equation 12), (Equation 15) to (Equation 17), (Equation 18) to (Equation 20), (Equation 21) to (Equation 24), ( 25), the estimated current Iue of the U phase, the estimated current Ive of the V phase, and the estimated current Iwe of the W phase are estimated.

That is, the estimated current calculator 24Ac calculates the induced voltages Eu, Ev, and Ew from the above (Equation 9) to (Equation 11) based on the induced voltage constant Ke, the electrical angle phase θe, and the electrical angle estimated speed ωe. do. Based on the output voltage command values Vu ^* , Vv ^* , Vw ^* and the induced voltages Eu, Ev, Ew, the estimated current calculator 24Ac calculates the neutral point potential Vn from the above (Equation 12). Based on the output voltage command values Vu ^* , Vv ^* , Vw ^* , the induced voltages Eu, Ev, Ew, and the neutral point potential Vn, the estimated current calculation unit 24Ac calculates the above (Equation 18) to (Equation 25). , the current amplitude correction coefficient Ka that does not include the Laplace operator s is calculated.

Based on the output voltage command values Vu ^* , Vv ^* , Vw ^* , the induced voltages Eu, Ev, Ew, the neutral point potential Vn, and the current amplitude correction coefficient Ka, the estimated current calculation unit 24Ac calculates the above (formula 15 ) to (Equation 17), the estimated current Iue of the U phase, the estimated current Ive of the V phase, and the estimated current Iwe of the W phase are estimated. Then, the estimated current calculation unit 23Ac outputs the estimated U-phase estimated current Iue, V-phase estimated current Ive, and W-phase estimated current Iwe to the current reproduction processing unit 24Ad.

The current reproduction processing unit 24Ad reproduces a three-phase current from the two-phase detection currents output from the detection current processing unit 24Ab based on the detection pattern output from the detection current phase/detection pattern determination processing unit 24Aa. and a process of reproducing a three-phase current from the one-phase detected current output from the detected current processing unit 24Ab and the estimated current output from the estimated current calculation unit 24Ac. The process of reproducing the three-phase current from the two-phase detected current output from the detected current processing section 24Ab is the same process as in the prior art.

The current reproduction processing unit 24Ad compares the detected current and the estimated in-phase current, and determines whether or not the error between the detected current and the estimated in-phase current is within a predetermined value (within the allowable error for normal detected current). do. That is, the current reproduction processing unit 24Ad determines the accuracy of the estimated current. If the error is within a predetermined value, the current reproduction processing unit 24Ad replaces the undetected two-phase current with the estimated current output from the estimated current calculation unit 24Ac. Alternatively, the current reproduction processing unit 24Ad may use the undetected current of one phase as the estimated current and calculate the remaining one phase from Kirchhoff's law.

If the error obtained by comparing the detected current and the in-phase estimated current exceeds a predetermined value, the current reproduction processing unit 24Ad determines that the estimated current contains a calculation error or that noise is superimposed on the detected current. Then, control is performed such that the detected current and the estimated current at this processing timing are not used for controlling the motor 1, for example, they are not output.

Alternatively, the error between the detected current and the estimated current obtained by comparing the detected current and the estimated in-phase current is used as a correction value for the undetected phase current, thereby improving the accuracy of the estimated current. becomes possible. For example, the error from the U-phase estimated current Iue when the U-phase detected current Iu can be detected is defined as the following (Equation 32).

ΔI=Iu−Iue (Formula 32)

According to Kirchhoff's law, the total value of the three-phase currents is 0, so the relationship between the estimated currents and the actual phase currents is given by the following (Equation 33).

Iu+Iv+Iw=Iue+Ive+Iwe (=0) (Equation 33)

By transforming (Equation 33) using the above (Equation 32), the following Equation (34) is obtained. The following (Formula 35) is obtained by transforming the following (Formula 34).

ΔI+Iue+Iv+Iw=Iue+Ive+Iwe (Formula 34)
Iv+Iw=Ive+Iwe−ΔI (Formula 35)

The total value of the undetectable phase current is corrected by using the error between the detected current and the estimated current of the phase that can be detected. , indicating that reproducibility is enhanced.

Iu+Iv+Iw=Iu+Ive+Iwe−ΔI (Formula 36)

In this manner, by comparing the detected current of the detected phase and the estimated current, it is possible to determine whether or not the accuracy of the estimated current is within the allowable error range. Therefore, the d-axis current Id and the q-axis current Iq, the U-phase output voltage command value Vu*, the V-phase output voltage command value Vv*, and the W-phase output voltage command required to calculate the estimated current at the next processing timing The amplitude component and the phase component obtained from the value Vw*, the estimated electrical angle speed ωe, and the electrical angle phase θe can be calculated with high accuracy. Therefore, even if the phase current of only one phase can be detected, the current can be accurately reproduced. In addition, if it is determined that the accuracy of the estimated current is within the allowable error range for the detected phase current of one phase, the accuracy of the estimated currents for the remaining two phases that have not been detected are also within the allowable error range. Since it can be estimated that it is within the range, it is possible to guarantee accuracy even for two-phase currents that cannot be detected, and the estimated current value becomes an accurate value.

<Current detection determination and d-axis current Id and q-axis current Iq at processing timing of the first embodiment>
FIG. 6 is a diagram for explaining current detection determination, d-axis current Id, and q-axis current Iq at processing timings. FIG. 6 shows the relationship between each processing timing (carrier frequency), current detection determination, d-axis current Id and q-axis current Iq. In FIG. 6, when the current detection determination is x (when the current detection is not possible), it is necessary to calculate the estimated current. At this time, since the d-axis current Id and the q-axis current Iq cannot be detected, the current amplitude (dq-axis current amplitude) Kdq on the two axes (dq-axis current amplitude) in the above (Equation 24) cannot be calculated. Therefore, an amplitude component (current amplitude correction coefficient Ka) that does not include the Laplace operator s cannot be calculated.

Therefore, as the d-axis current Id and the q-axis current Iq at the processing timing when the current cannot be detected, the current amplitude (dq-axis current amplitude) Kdq is obtained. Alternatively, the current amplitude correction coefficient Ka itself may be complemented before and after the processing timing at which the current could not be detected by a method such as filtering or interpolation (or extrapolation).

(Modification of Embodiment 1)
In the first embodiment, the d-axis current Id and the q-axis current Iq at the processing timing when the current could not be detected are complemented by a method such as a filter or interpolation (or extrapolation), and the current amplitude (dq-axis current amplitude) Kdq is obtained. I asked for it. However, not limited to this, the 3φ/dq converter 25 may have a storage unit (not shown) that stores the previous values of the d-axis current Id and the q-axis current Iq at the processing timing when the current was detected. . In this case, the 3φ current calculator 24A outputs a signal indicating that current detection is impossible to the 3φ/dq converter 25 at the processing timing when the current detection is not possible. 3φ/dq converter 25 converts the previous values of d-axis current Id and q-axis current Iq stored in a storage unit (not shown) to 3φ Output to the current calculator 24A. Note that the storage unit (not shown) that stores the previous values of the d-axis current Id and the q-axis current Iq is not limited to being included in the 3φ/dq converter 25. It may be provided upstream of the processor 26 independently of the 3φ/dq converter 25 .

Since the first embodiment described above uses the current estimating means that does not include the Laplace operator s, it is possible to lighten the calculation load of the estimated current. Moreover, since Embodiment 1 uses current estimation means that does not include various current-dependent and temperature-dependent parameters, it is possible to improve the accuracy of current estimation. Further, in the first embodiment, it is determined whether or not the accuracy of the estimated current is within the allowable range, and the estimated current is error-corrected according to the determination result, so the current can be reproduced with high accuracy.

[Embodiment 2]
<Theoretical Background of Embodiment 2>
(Overview of Embodiment 2)
In the first embodiment described above, as shown in FIG. 1, the induced voltage of each phase of the U-phase, V-phase, and W-phase of the motor 1 when the motor 1 is replaced by an equivalent circuit is calculated using the induced voltage constant Ke. Then, the estimated induced voltage is used to estimate the current of each of the U-phase, V-phase, and W-phase of the motor 1 . Here, when calculating the estimated current from the induced voltage, if the induced voltage constant Ke, the rotation speed (rotational speed) ω, and the rotor rotation angle (phase) θ, which are motor parameters, are not correct, the current cannot be estimated with high accuracy. do not have. In addition, the induced voltage constant Ke affects the accuracy of the estimated current due to factors such as individual variations in motors, temperature dependence of magnets in motors, and deterioration over time.

Specifically, the above (Equation 9) to (Equation 11) contain errors because the values of the induced voltages Eu, Ev, and Ew estimated for each phase fluctuate due to changes in the induced voltage constant Ke. The estimated currents of (Equation 15) to (Equation 17) calculated using the induced voltages Eu, Ev, and Ew also contain errors, and the estimation accuracy of the estimated currents is lowered. Therefore, in the second embodiment, the induced voltages Eu, Ev, and Ew of the U-phase, V-phase, and W-phase of the motor 1 are calculated using the d-axis voltage command value Vd ^* and the q-axis voltage command value without using the induced voltage constant Ke. By obtaining from the value Vd ^* , the calculation accuracy of the induced voltage is made better than the estimation.

Specifically, the induced voltages Eu, Ev, and Ew of the three phases U, V, and W are calculated from the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* .

A method of calculating the induced voltages Eu, Ev, and Ew of the three phases U, V, and W from the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq of the second embodiment will be described below. FIG. 7 is a basic vector diagram in the motor model. A motor model formula is shown in (Formula 37) below. FIG. 7 illustrates the following motor model equation (Equation 37) ignoring the transient term. In FIG. 7, "Vo" is the induced voltage.

Also, FIG. 8 is a diagram for explaining the calculation of the induced voltage on the dq axis. As shown in FIG. 8, the induced voltage Vod on the d-axis is a voltage Ra· It is obtained by subtracting the Id. The resistance value Ra is the resistance value of the motor 1 for each phase of the d-axis and the q-axis. The induced voltage Voq on the q-axis is obtained by subtracting the voltage Ra·Iq corresponding to the product of the q-axis current Iq and the resistance value Ra from the q-axis component Vq of the dq-axis voltage Va. This relationship is shown in (Equation 38) and (Equation 39) below.

Vod=Vd ^* -Ra·Id (Equation 38)
Voq=Vq ^* -Ra·Iq (Formula 39)

From (Eq.38), (Eq.39), FIGS. 7 and 8, it can be seen that the dq-axis voltage Va can be obtained from the induced voltage Vo, the dq-axis current Ia (see FIG. 7), and the resistance value Ra. The induced voltage constant Ke is obtained from the interlinkage magnetic flux Ψa. coincide with each other, it can be seen that the value obtained by multiplying the combined magnetic flux linkage Ψo by the rotation speed (rotational speed) ω is the induced voltage Vo.

The induced voltage Vod on the d-axis and the induced voltage Voq on the q-axis thus obtained are converted into induced voltages Eu, Ev, and Ew of three phases of U-phase, V-phase, and W-phase. The 2-phase to 3-phase conversion is performed in the same manner as the conversion from the d-axis voltage command value and the q-axis voltage command value to the output voltage command values (applied voltages) Vu ^* , Vv ^* , Vw ^* which are normally performed. As a result, the induced voltage can be calculated without using the induced voltage constant Ke whose value varies with changes in the operating environment. Calculation formulas for calculating the three-phase induced voltages Eu, Ev, and Ew from the two-phase induced voltage Vod and the induced voltage Voq are shown in the following (Equation 40) to (Equation 42).

Eu=√(2/3)·(Vod·cos θ−Voq·sin θ) (Formula 40)
Ev=√(2/3)·{Vod·(√(3)·sin θ−cos θ)/2
+Voq (sin θ + √(3) cos θ)/2) (Equation 41)
Ew=√(2/3)·{Vod·(−√(3)·sin θ−cos θ)/2
+Voq (sin θ−√(3) cos θ)/2) (Equation 42)

By directly obtaining the induced voltages Eu, Ev, and Ew from the d-axis voltage command value Vd ^* and the q-axis voltage command value Vq ^* as in the above (Equation 38) to (Equation 42), the accuracy is higher than that of the conventional technique. high induced voltages Eu, Ev, and Ew can be calculated. As a result, the estimation accuracy of the estimated currents Iue, Ive, Iwe calculated using the above (Equation 15) to (Equation 17) can be improved.

<Regarding the configuration of the motor control device according to the second embodiment>
FIG. 9 is a diagram showing an example of the configuration of a motor control device according to the second embodiment. Compared to the motor control device 100A of the first embodiment, the motor control device 100B of the second embodiment has a microcomputer 10B instead of the microcomputer 10A, and a 3φ current calculator 24A that is part of the microcomputer 10A. has a 3φ current calculator 24B.

dq/3φ converter 21 outputs U-phase output voltage command value Vu ^* , V-phase output voltage command value Vv ^* and W-phase output voltage command value Vw ^* to PWM generator 22 and 3φ current calculator 24B. Further, the 3φ/dq converter 25 transfers the d-axis current Id to the subtractor 14, the non-interfering controller 18, the 3φ current calculator 24B, the axis error arithmetic processor 26, the q-axis current Iq to the subtractor 15, the Output to the interfering controller 18, the 3φ current calculator 24B, and the axis error processor 26, respectively.

Further, the subtractor 19 subtracts the d-axis decoupling correction value Vda output from the decoupling controller 18 from the d-axis voltage command value Vd ^** output from the d-axis current controller 16 to obtain d A d-axis voltage command value Vd ^* is generated by decoupling the axis voltage command value Vd ^** , and is output to the dq/3φ converter 21, the 3φ current calculator 24B, and the axis error processor 26. The adder 20 adds the q-axis decoupling correction value Vqa output from the decoupling controller 18 to the q-axis voltage command value Vq** output from the q-axis current controller 17 to obtain the q-axis voltage A q-axis voltage command value Vq* is generated by decoupling the command value Vq**, and is output to the dq/3φ converter 21, the 3φ current calculator 24B, and the axis error processor 26.

In addition, the 1-shunt current detector 32 receives information on the detected bus current and switching information on the 6-phase PWM signals (Up, Vp, Wp, Un, Vn, Wn) output from the PWM generator 22, Output to the 3φ current calculator 24B.

<Regarding the configuration of the 3φ current calculator of the second embodiment>
FIG. 10 is a diagram showing an example of the configuration of a 3φ current calculator according to the second embodiment. As shown in FIG. 10, the 3φ current calculator 24B of the second embodiment has a detected current phase/detected pattern determination processing section 24Ba, a detected current processing section 24Bb, an estimated current calculation section 24Bc, and a current reproduction processing section 24Bd. Compared to the 3φ current calculator 24A of the first embodiment, the 3φ current calculator 24B differs from the estimated current calculator 24Ac only in an estimated current calculator 24Bc, and includes a detected current phase/detection pattern determination processor 24Ba and a detected current process. The section 24Bb and the current reproduction processing section 24Bd are the same as the detected current phase/detection pattern determination processing section 24Aa, the detected current processing section 24Ab, and the current reproduction processing section 24Ad.

<Regarding the configuration of the estimated current calculator of the second embodiment>
FIG. 11 is a diagram illustrating an example of the configuration of an estimated current calculator according to the second embodiment; The estimated current calculator 24Bc of the second embodiment has a dq-axis induced voltage calculator 24Bc-1, a 3φ-axis induced voltage calculator 24Bc-2, a current amplitude calculator 24Bc-3, and an estimated current calculator 24Bc-4.

Based on the d-axis current Id, the q-axis current Iq, the d-axis voltage command value Vd ^* , the q-axis voltage command value Vq ^* , and the motor parameter Ra, the dq-axis induced voltage calculator 24Bc-1 calculates the above (Equation 38). The induced voltage Vod on the d-axis and the induced voltage Voq on the q-axis are calculated from (Equation 39). Then, the dq-axis induced voltage calculator 24Bc-1 outputs the calculated induced voltage Vod on the d-axis and the induced voltage Voq on the q-axis to the 3φ-axis induced voltage calculator 24Bc-2. Note that the dq-axis induced voltage calculator 24Bc-1 uses the resistance value Ra of the motor 1, which is given from outside such as a storage device, as the motor parameter "Ra" in the above (Equation 38) to (Equation 39). .

Based on the induced voltage Vod on the d-axis, the induced voltage Voq on the q-axis and the electrical angle phase θe, the 3φ-axis induced voltage calculator 24Bc-2 calculates the 3φ-axis The above induced voltages Eu, Ev, and Ew are calculated. Then, the 3φ-axis induced voltage calculator 24Bc-2 outputs the calculated induced voltages Eu, Ev, and Ew on the 3φ-axis to the estimated current calculator 24Bc-4. The 3φ-axis induced voltage calculator 24Bc-2 uses the electrical angle phase θe output from the position estimator 30 as “θ” in the above (Equation 40) to (Equation 42).

Based on the d-axis current Id and the q-axis current Iq, the current amplitude calculation unit 24Bc-3 calculates the current amplitude (dq-axis current amplitude) Kdq on the two axes from the above (Equation 24), and calculates the estimated current. Output to the section 24Bc-4.

The estimated current calculator 24Bc-4 calculates the induced voltages Eu, Ev, and Ew on the 3φ axes, the current amplitude on the 2 axes (dq-axis current amplitude) Kdq, the U-phase output voltage command value Vu ^* , and the V-phase output voltage command value. Based on Vv ^* and the W-phase output voltage command value Vw ^* , the above (Equation 14), (Equation 15) to (Equation 17), (Equation 18) to (Equation 20), (Equation 21) to (Equation 22) ), (Equation 23) to (Equation 24), and (Equation 25), the three-phase estimated currents Iue, Ive, and Iwe are calculated. Then, the estimated current calculator 24Bc-4 outputs the calculated three-phase estimated currents Iue, Ive, and Iwe to the current reproduction processor 24Bd.

According to the second embodiment described above, the induced voltages Eu, Ev, and Ew on the 3φ axis are obtained from the U-phase output voltage command value Vu ^* , the V-phase output voltage command value Vv ^* , and the W-phase output voltage command value Vw ^* . . As a result, the induced voltages Eu, Ev, and Ew can be accurately calculated in real time without using the induced voltage constant Ke, which is affected by individual variations in the motor 1, temperature dependence of the magnets of the motor 1, and deterioration over time. Therefore, the estimated currents Iue, Ive, and Iwe of the three phases of the induced voltage can be estimated with higher accuracy.

The specific names, processes, controls, and information including various data and parameters in the above-described embodiments and illustrations are only examples, and can be changed as appropriate unless otherwise specified. Also, the configuration of each unit or each device in the above-described embodiments may be appropriately distributed or integrated in consideration of the processing load, implementation efficiency, and the like. Further, each process in the above-described embodiments may be executed by appropriately changing the order of the processes in consideration of the processing load, implementation efficiency, and the like.

The broader aspects of the above-described embodiments are not limited to the specific details and representative embodiments shown and described above. Accordingly, various changes may be made without departing from the general inventive concept or scope as defined by the appended claims and equivalents thereof.

1 motors 10A, 10B microcomputer 11 subtractor 12 speed controller 13 excitation current controller 14 subtractor 15 subtractor 16 d-axis current controller 17 q-axis current controller 18 decoupling controller 19 subtractor 20 adder 21 dq/3φ converter 22 PWM generator 23 IPM
24A, 24B 3φ current calculators 24Aa, 24Ba Detected current phase/detected pattern determination processors 24Ab, 24Bb Detected current processors 24Ac, 24Bc Estimated current calculators 24Ad, 24Bd Current reproduction processors 24Bc-1 dq axis induced voltage calculator 24Bc -2 3φ axis induced voltage calculation unit 24Bc-3 current amplitude calculation unit 24Bc-4 estimated current calculation unit 25 3φ/dq converter 26 axis error calculation processor 29 PLL controller 30 position estimator 31 1/Pn processor 32 1 Shunt current detectors 100A, 100B Motor control device

Claims (1)

a driving unit for driving the motor by supplying an applied voltage generated based on the difference between the target speed and the current speed of the motor to the motor;
a detection unit for detecting a phase current of each phase in the UVW coordinate system flowing through the motor;
a generation unit that generates the applied voltage of each phase in the UVW coordinate system from the applied voltage of each phase in the dq coordinate system,
The detection unit is
a current estimator for estimating a phase current of each phase in the UVW coordinate system flowing through the motor;
Among the phases in the UVW coordinate system flowing through the motor, the detected phase current is output for the phase current that could be detected, and the phase current estimated by the current estimator is output for the phase that could not be detected. and a current reproduction processing unit,
The current estimator calculates the induced voltage of each phase in the UVW coordinate system of the motor based on the voltage command value of each phase in the dq coordinate system, and calculates the phase current of each phase in the UVW coordinate system that flows through the motor. is estimated based on a voltage value obtained by subtracting the neutral point potential of the motor and the induced voltage from the applied voltage of the motor, and the applied voltage and the induced voltage.
motor controller.

Images