Control device

Abstract

An object of the present invention is to provide a control device capable of reducing a magnetic pole position estimation error caused by an interference inductance.SOLUTION: The control device 1 is a control device that estimates and controls a magnetic pole position by superimposing a high-frequency voltage on a synchronous motor 3 having a rotor with saliency. This device is provided with a high frequency current extracting part 21 for applying a high frequency voltage on an estimated d-axis and separating a high frequency current detected as its response into an estimated d-axis component Idch and an estimated q-axis component Iqch and an inductance reference part for providing a reference inductance value including a d-d axis self-inductance value Ldd, a q-q axis self-inductance value Lqq, a d-q axis interference inductance value and a q-d axis interference inductance value according to a torque command value Tref, a current command value Iid_ref, Iiq_ref or a detected current value and a magnetic pole position error estimating part 2 for estimating a magnetic pole position error estimated value Δθc by using the estimated d-axis component Idch, the estimated q-axis component Iqch and the reference inductance value.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a control device for a synchronous motor. [Background technology]

[0002] The control device must apply power at the appropriate timing depending on the rotor's magnetic pole position. The rotor position can be measured using a rotational position sensor, but this is expected to complicate the structure, increase costs, and increase the risk of failure, so sensorless control, which does not use a rotational position sensor, has been proposed (for example, Patent Documents 1-3). [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 5396876 [Patent Document 2] Patent No. 6551473 [Patent Document 3] Patent No. 5652664 Summary of the Invention [Problem to be solved by the invention]

[0004] One method of sensorless control is a high-frequency superposition sensorless magnetic pole position estimation method, in which a voltage is applied at a frequency higher than the power frequency that generates the rotational force, and the magnetic pole position is estimated from the current response.

[0005] However, in the sensorless magnetic pole position estimation method using high frequency superposition, voltage and current equations that are free from interference between the d and q axes are used to calculate the magnetic pole position error, but in motors with a structure in which interference inductance between the d and q axes is large, the magnetic pole position error accuracy may deteriorate (for example, Patent Document 1).

[0006] To solve the above problem, there are sensorless magnetic pole position estimation methods that take into account the interference inductance between the d and q axes. One example is a technology that calculates an interference coefficient that correlates with the mutual interference between the d and q axes and uses it to calculate the magnetic pole position, but the interference coefficient is calculated from past angle estimates and response values ​​(for example, Patent Document 2). Another example is a method that corrects the magnetic pole position estimate, which is estimated from a high-frequency voltage command, which is a response component when a high-frequency current is applied, based on magnetic flux interference error (Patent Document 3).

[0007] However, these sensorless magnetic pole position estimation methods have room for improvement in terms of suppressing the influence of magnetic pole position estimation errors caused by interference inductance between the d and q axes and improving estimation accuracy.

[0008] SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned problems, and an object of the present invention is to provide a control device capable of reducing magnetic pole position estimation errors caused by interference inductance. [Means for solving the problem]

[0009] In order to solve the above-described problems, a control device according to one embodiment of the present invention is a control device that estimates and controls a magnetic pole position by superimposing a high-frequency voltage on a synchronous motor having a rotor with salient poles, and includes: a high-frequency current extraction unit that applies a high-frequency voltage to the estimated d-axis when an orthogonal coordinate system is defined with the d-axis representing the easy direction of magnetization and the q-axis representing the hard direction of magnetization, and separates the high-frequency current detected in response to the application of the high-frequency voltage into an estimated d-axis component and an estimated q-axis component; an inductance reference unit that provides reference inductance values ​​including a d-axis self-inductance value, a q-axis self-inductance value, a dq-axis interference inductance value, and a qd-axis interference inductance value according to a torque command value, a current command value, or a detected current value; and a magnetic pole position error estimator that estimates a magnetic pole position error using the estimated d-axis component, the estimated q-axis component, and the reference inductance value.

[0010] Any combination of the above components, or mutual substitution of the components or expressions of the present invention between methods, systems, etc., are also valid aspects of the present invention. [Effects of the Invention]

[0011] According to the present invention, it is possible to provide a control device capable of reducing magnetic pole position estimation errors caused by interference inductance. [Brief explanation of the drawings]

[0012] [Figure 1] 1 is a block diagram showing an electric motor system including a control device according to an embodiment; [Figure 2] 2 is a timing chart showing the operation of the high-frequency current extraction unit of the control device of FIG. [Figure 3] 2 is a block diagram schematically showing a magnetic pole position error estimating unit of the control device of FIG. 1. FIG. [Figure 4] FIG. 1 is a diagram showing a schematic diagram of the structural d-axis and q-axis and the estimated d-axis and q-axis. [Figure 5] 2 is a block diagram schematically showing a magnetic pole position calculation unit of the control device of FIG. 1. FIG. DETAILED DESCRIPTION OF THE INVENTION

[0013] The present invention will be described below based on preferred embodiments with reference to the drawings. In the embodiments and modifications, identical or equivalent components and members are designated by the same reference numerals, and redundant explanations will be omitted where appropriate. The dimensions of the members in the drawings are enlarged or reduced as appropriate to facilitate understanding. Some members that are not important for explaining the embodiments will be omitted from the drawings.

[0014] Furthermore, terms including ordinal numbers such as first and second are used to describe various components, but these terms are used only to distinguish one component from another and do not limit the components.

[0015] First, the background to the invention will be explained. When vector control is performed by a sensorless control method using saliency in a synchronous reluctance motor that does not use a permanent magnet in the rotor, the effect of inductance change on magnetic pole position estimation error is large due to the characteristic of large magnetic saturation, which is a factor in reducing the accuracy of magnetic pole position estimation.

[0016] The inventors have focused on the interference inductance value and have found that by estimating the magnetic pole position error by taking the interference inductance value into consideration, it is possible to suppress the influence of inductance changes on the magnetic pole position estimation error even when the amount of current flow increases, strengthening the influence of magnetic saturation and resulting in nonlinear characteristics. The present invention has been made based on this finding, and will be described below with reference to the embodiments.

[0017] [Embodiment] The configuration of a control device 1 according to an embodiment will be described with reference to FIGS. 1 to 5. FIG. 1 is a block diagram illustrating an example of an electric motor system 100 including the control device 1 according to the embodiment. The electric motor system 100 includes an electric motor 3 and a control device 1 that controls the electric motor 3. The electric motor 3 is a three-phase synchronous motor having a rotor 4 with salient poles. As an example, the electric motor 3 is a synchronous reluctance motor that does not use a permanent magnet in the rotor 4. The control device 1 mainly includes an inverter unit 2, a PWM generation unit 5, a phase current detection unit 6, a first coordinate transformation unit 7, a current control unit 8, a current command unit 9, a second coordinate transformation unit 10, a high-frequency voltage command generation unit 11, and a magnetic pole position estimation unit 20. The magnetic pole position estimation unit 20 includes a high-frequency current extraction unit 21, a magnetic pole position error estimation unit 22, and a magnetic pole position calculation unit 23.

[0018] The functional blocks in the block diagrams shown in Figures 1, 3, and 5 can be realized in terms of hardware by computer processors, CPUs, memory, and other elements, electronic circuits, and mechanical devices, and in terms of software by computer programs, etc. However, the functional blocks shown here are realized by the cooperation of these elements. Therefore, it will be understood by those skilled in the art that these functional blocks can be realized in various ways by combining hardware and software.

[0019] Hereinafter, a Cartesian coordinate system is defined in which the easy direction of magnetization is the d-axis and the hard direction of magnetization is the q-axis. In this embodiment, the magnetic pole position error estimated value Δθc is estimated using a rectangular high-frequency voltage superposition method. In this method, the high-frequency voltage command generator 11 superimposes a high-frequency voltage Vh on the d-axis voltage Vd output from the current controller 8. The high-frequency voltage Vh is a rectangular high-frequency voltage with a frequency sufficiently higher than the drive frequency of the electric motor 3. The frequency of the high-frequency voltage Vh is not particularly limited, but is preferably at least five times the power frequency at which the electric motor 3 generates torque and not more than one-fifth the PWM frequency.

[0020] The high-frequency components of the high-frequency voltage Vh superimposed on the d-axis voltage Vd are included in the d-axis current Id and the q-axis current Iq via the second coordinate converter 10, the PWM generator 5, the inverter 2, and the first coordinate converter 7. The amplitudes of the high-frequency components included in the d-axis current Id and the q-axis current Iq change based on the magnetic pole position of the motor 3. In other words, the magnetic flux density of the salient pole of each phase changes depending on the magnetic pole position, and accordingly, the amplitudes of the high-frequency components included in the d-axis current Id and the q-axis current Iq change due to the nonlinear characteristics of the magnetic permeability caused by magnetic saturation of the salient pole of each phase.

[0021] By applying the above concept, the magnetic pole position estimator 20 estimates the magnetic pole position of the electric motor 3 based on the amplitude of the high-frequency components contained in the d-axis current Id and the q-axis current Iq. The magnetic pole position estimator 20 includes a high-frequency current extractor 21, a magnetic pole position error estimator 22, and a magnetic pole position calculator 23.

[0022] The d-axis current Id and the q-axis current Iq are input to the high-frequency current extraction unit 21. The high-frequency current extraction unit 21 applies a high-frequency voltage to the estimated d-axis, and separates the high-frequency current detected as a response thereto into an estimated d-axis component Idch and an estimated q-axis component Iqch.

[0023] A specific example of the high-frequency current extraction unit 21 will be described with reference to FIG. 2 as well. FIG. 2 is a timing chart showing the operation of the high-frequency current extraction unit 21. In this figure, a rectangular wave represents the high-frequency voltage waveform g1, and a triangular wave represents the high-frequency current waveform g2. The high-frequency current extraction unit 21 extracts a high-frequency current that fluctuates in accordance with the high-frequency voltage Vh from the d-axis current Id and the q-axis current Iq. The extracted high-frequency current has a triangular waveform as shown by g2 in FIG. 2. The estimated d-axis component Idch and the estimated q-axis component Iqch are fluctuation values ​​of the high-frequency current that fluctuate in accordance with the high-frequency voltage Vh shown by g1, and these fluctuation values ​​are specified as the difference between the peak value and the trough value of the extracted high-frequency current. The trough value of the high-frequency current appears at the timing of the rising edge of the high-frequency voltage Vh, and the peak value appears at the timing of the falling edge of the high-frequency voltage Vh. The estimated d-axis component Idch is the difference between the peak value and the bottom value of the d-axis current Id, and the estimated q-axis component Iqch is the difference between the peak value and the bottom value of the q-axis current Iq.

[0024] The magnetic pole position error estimating unit 22 will be described with reference to Fig. 3. Fig. 3 is a block diagram schematically showing an example of the magnetic pole position error estimating unit 22. The magnetic pole position error estimating unit 22 includes an inductance reference unit 24 and a calculation unit 25.

[0025] As an example, the inductance reference unit 24 may provide a reference inductance value Laa in response to the detected d-axis current Id and q-axis current Iq.

[0026] The inductance reference unit 24 stores a reference inductance value Laa including the d-axis self-inductance value Ldd, the q-axis self-inductance value Lqq, the dq-axis interference inductance value Ldq, and the qd-axis interference inductance value Lqd. The reference inductance value Laa may be information acquired in advance or information measured at a predetermined timing. In this embodiment, the inductance reference unit 24 provides the reference inductance value Laa according to the detected current values ​​Id and Iq. Hereinafter, the dq-axis interference inductance value Ldq and the qd-axis interference inductance value Lqd may be collectively referred to simply as "interference inductance values."

[0027] Here, the dq-axis interference inductance value Ldq is a value that indicates the influence that magnetic flux changes based on current changes on the d-axis have on magnetic flux changes on the q-axis, and the qd-axis interference inductance value Lqd is a value that indicates the influence that magnetic flux changes based on current changes on the q-axis have on magnetic flux changes on the d-axis.

[0028] As another example, the inductance reference unit 24 may provide the reference inductance value Laa in accordance with the d-axis component and q-axis component of the current command value. The d-axis component of the current command value is the current command value Id_ref, and the q-axis component is Iq_ref. The current command values ​​Id_ref and Iq_ref will be described later.

[0029] In other words, since the detected current values ​​Id, Iq are correlated with the current command values ​​Id_ref, Iq_ref and the torque command value Tref, the inductance reference unit 24 may be configured to provide a reference inductance value Laa corresponding to the current command values ​​Id_ref, Iq_ref or the torque command value Tref instead of the detected current values ​​Id, Iq.

[0030] According to the inventor's investigations, it has been found that the magnetic pole position error estimated value Δθc is affected by the estimated d-axis component Idch, the estimated q-axis component Iqch, the d-axis self-inductance value Ldd, the q-axis self-inductance value Lqq, the dq-axis interference inductance value Ldq, and the qd-axis interference inductance value Lqd. Therefore, the magnetic pole position error estimator 22 estimates the magnetic pole position error estimated value Δθc using the estimated d-axis component Idch, the estimated q-axis component Iqch, and the reference inductance value Laa.

[0031] As an example, the magnetic pole position error estimating unit 22 may estimate the magnetic pole position error estimated value Δθc by linear calculation of the estimated d-axis component Idch and the estimated q-axis component Iqch.

[0032] For example, the calculation unit 25 of the magnetic pole position error estimating unit 22 calculates the magnetic pole position error estimated value Δθc by substituting the estimated d-axis component Idch, the estimated q-axis component Iqch, and the reference inductance value Laa into a set of simultaneous equations that can be set by experiment or simulation.

[0033] The magnetic pole position calculation unit 23 will be described with reference to Figures 4 and 5. Figure 4 is a diagram showing the structural d-axis and q-axis, and the estimated dc-axis and qc-axis. In this figure, the longitudinal direction of the rotor schematic diagram indicates the direction of easy magnetization, and the lateral direction indicates the direction of hard magnetization. The structural d-axis is an axis extending in the direction of easy magnetization determined by the rotor shape and structure, and the d-axis and q-axis are perpendicular to each other.

[0034] At low currents, the effect of magnetic saturation is small and no nonlinear characteristics are observed, but as the amount of current increases, the effect of magnetic saturation becomes stronger, and nonlinear characteristics are thought to occur. In the example of Figure 4, the estimated dc-axis and qc-axis are rotated clockwise from the structural d-axis and q-axis by the estimated magnetic pole position error Δθc.

[0035] 5 is a block diagram showing a schematic example of the magnetic pole position calculation unit 23. The magnetic pole position calculation unit 23 calculates the estimated magnetic pole position θe,Low by a PLL that feeds back the input magnetic pole position error estimated value Δθc to zero. The estimated magnetic pole position θe,Low is provided to the first coordinate conversion unit 7 and the second coordinate conversion unit 10.

[0036] Other configurations of the control device 1 will be described mainly with reference to Fig. 1. A current command unit 9 provides a d-axis current command value Id_ref and a q-axis current command value Iq_ref in accordance with, for example, a torque command value Tref for the electric motor system 100. Note that the d-axis current command value Id_ref and the q-axis current command value Iq_ref may be collectively referred to as the current command values ​​Id_ref and Iq_ref.

[0037] The control device 1 controls the drive current of the electric motor 3 based on the current command values ​​Id_ref and Iq_ref provided from the current command unit 9. DC power is supplied to the inverter unit 2 of the electric motor system 100 from a power supply means (not shown). A predetermined DC power is supplied to the electric circuits of the electric motor system 100 other than the inverter unit 2 from a power supply means (not shown).

[0038] Each phase output terminal of the three-phase inverter unit 2 is connected to one end of a corresponding phase winding (not shown) of the electric motor 3. Each phase output terminal of the inverter unit 2 is connected to a corresponding phase input terminal of the phase current detection unit 6. The phase currents Iu, Iv, and Iw detected by the phase current detection unit 6 are input to a first coordinate conversion unit 7. The three-phase currents input to the first coordinate conversion unit 7 are converted into a d-axis current Id and a q-axis current Iq by a three-phase / dq conversion.

[0039] The converted d-axis current Id and q-axis current Iq are input to a current control unit 8. A d-axis current command value Id_ref and a q-axis current command value Iq_ref generated by a current command unit 9 are input to the current control unit 8. The current control unit 8 amplifies the differences between the current command values ​​Id_ref, Iq_ref and the currents Id, Iq using an error amplifier 82 with a predetermined transfer function, thereby generating a d-axis voltage Vd and a q-axis voltage Vq. In other words, the current control unit 8 generates a d-axis voltage Vd and a q-axis voltage Vq according to the differences between the current command values ​​Id_ref, Iq_ref and the currents Id, Iq.

[0040] The d-axis voltage Vd and the q-axis voltage Vq are input to a second coordinate conversion unit 10, which converts them into three-phase voltages Vu, Vv, and Vw by dq / three-phase conversion. The converted three-phase voltages Vu, Vv, and Vw are input to a PWM generation unit 5. Based on the three-phase voltages Vu, Vv, and Vw, the PWM generation unit 5 generates PWM signals Vu+, Vu-, Vv+, Vv-, Vw+, and Vw- for driving switching elements of each phase arm that constitutes the inverter unit 2. The PWM signals Vu+, Vu-, Vv+, Vv-, Vw+, and Vw- are input to the inverter unit 2.

[0041] The d-axis current Id and q-axis current Iq provided from the first coordinate conversion unit 7 are input to the magnetic pole position estimation unit 20. As described above, the magnetic pole position estimation unit 20 estimates the magnetic pole positions θe, Low of the electric motor 3 based on the input d-axis current Id and q-axis current Iq, and provides them to the first coordinate conversion unit 7 and the second coordinate conversion unit 10. The first coordinate conversion unit 7 and the second coordinate conversion unit 10 perform coordinate axis transformation using the magnetic pole positions θe, Low.

[0042] The operation of the control device 1 of this embodiment configured as described above will be described. Based on the provided current command values ​​Id_ref and Iq_ref, a drive current flows to each phase of the electric motor 3 via the second coordinate conversion unit 10, the PWM generation unit 5, and the inverter unit 2, causing the rotor 4 to rotate. The phase currents Iu, Iv, and Iw detected by the phase current detection unit 6 are converted to currents Id and Iq by the first coordinate conversion unit 7 and fed back to the current control unit 8. This feedback control controls the phase currents Iu, Iv, and Iw of the electric motor 3 to follow the current command values ​​Id_ref and Iq_ref. The magnetic pole position estimation unit 20, which includes the high-frequency current extraction unit 21, the magnetic pole position error estimator 22, and the magnetic pole position calculation unit 23, estimates the magnetic pole positions θe and Low based on the currents Id and Iq and provides them to the first coordinate conversion unit 7 and the second coordinate conversion unit 10.

[0043] The features of the control device 1 of the embodiment configured as above will be described below. The control device 1 is a control device that estimates and controls the magnetic pole position by superimposing a high frequency voltage on a synchronous motor 3 having a rotor 4 with salient poles. When an orthogonal coordinate system is defined with the d-axis representing the easy direction of magnetization and the q-axis representing the hard direction of magnetization, the magnetic pole position error estimation unit 22 includes: a high-frequency current extraction unit 21 that applies a high-frequency voltage to the estimated d-axis and separates the high-frequency current detected in response thereto into an estimated d-axis component Idch and an estimated q-axis component Iqch; an inductance reference unit 24 that provides a reference inductance value Laa including a d-axis self-inductance value Ldd, a q-axis self-inductance value Lqq, a dq-axis interference inductance value Ldq, and a qd-axis interference inductance value Lqd in accordance with a torque command value Tref, a current command value Id_ref, Iq_ref, or the detected current values ​​Id, Iq; and a magnetic pole position error estimation unit 22 that estimates a magnetic pole position error estimated value Δθc using the estimated d-axis component Idch, the estimated q-axis component Iqch, and the reference inductance value Laa.

[0044] This configuration allows for highly accurate sensorless control that is less susceptible to the influence of interference inductance, even when the amount of current flow increases, causing magnetic saturation and resulting in nonlinear characteristics. As a result, a control device can be provided that can reduce magnetic pole position estimation errors caused by interference inductance.

[0045] The above describes in detail examples of embodiments of the present invention. The above-described embodiments merely illustrate specific examples of implementing the present invention. The contents of the embodiments do not limit the technical scope of the present invention, and many design changes, such as changes, additions, and deletions of components, are possible within the scope of the invention as defined in the claims. In the above-described embodiments, content that allows such design changes is described using notations such as "in the embodiment" or "in the embodiment," but design changes are also permissible in content that does not have such notations.

[0046] (Variation) The following describes the modified examples. In the drawings and descriptions of the modified examples, the same or equivalent components and members as those in the embodiment are denoted by the same reference numerals. Explanations that overlap with the embodiment will be omitted as appropriate, and the description will focus on the configurations that differ from the embodiment.

[0047] In the above description, an example has been shown in which the control device 1 includes the current command unit 9 and uses the torque command value Tref, but the present invention is not limited to this. For example, the control device 1 may be configured not to include the current command unit 9 or not to use the torque command value Tref.

[0048] In the above description, an example has been shown in which the magnetic pole position error estimating unit 22 estimates the magnetic pole position error estimated value Δθc by linear calculation, but the present invention is not limited to this.

[0049] For example, the magnetic pole position error estimating unit 22 may be configured to input the estimated d-axis component Idch and the estimated q-axis component Iqch into a lookup table created in advance for the reference inductance value Laa, and provide the magnetic pole position error estimated value Δθc as an output value of the lookup table.

[0050] For example, the magnetic pole position error estimating unit 22 may be configured to input the reference inductance value Laa, the estimated d-axis component Idch, and the estimated q-axis component Iqch into a learning model generated using past measurement data, and provide the magnetic pole position error estimated value Δθc as the output value of the learning model.The learning model may be an AI model generated by machine learning using past measurement data of the reference inductance value Laa, the estimated d-axis component Idch, and the estimated q-axis component Iqch, and the magnetic pole position error estimated value Δθc.The learning model may be generated using a known machine learning method, such as a support vector machine, a neural network (including deep learning), or a random forest.

[0051] Each of these modifications provides the same functions and effects as the embodiment.

[0052] Any combination of the above-described embodiments and modifications is also useful as an embodiment of the present invention. A new embodiment resulting from the combination has the combined effects of the combined embodiments and modifications. [Explanation of symbols]

[0053] 1 control device, 2 inverter unit, 3 electric motor, 4 rotor, 5 PWM generation unit, 6 phase current detection unit, 7 first coordinate transformation unit, 8 current control unit, 9 current command unit, 10 second coordinate transformation unit, 11 high frequency voltage command generation unit, 20 magnetic pole position estimation unit, 21 high frequency current extraction unit, 22 magnetic pole position error estimation unit, 23 magnetic pole position calculation unit, 24 inductance reference unit, 25 calculation unit, 82 error amplifier, 100 electric motor system.

Claims

1. A control device that estimates and controls a magnetic pole position by superimposing a high frequency voltage on a synchronous motor having a rotor with salient poles, a high frequency current extracting unit that applies a high frequency voltage to an estimated d-axis when a Cartesian coordinate system is defined with the easy magnetization direction as the d-axis and the hard magnetization direction as the q-axis, and separates a high frequency current detected as a response thereto into an estimated d-axis component and an estimated q-axis component; an inductance reference unit that provides reference inductance values ​​including a d-axis self-inductance value, a q-axis self-inductance value, a d-q axis interference inductance value, and a q-d axis interference inductance value according to a torque command value, a current command value, or a detected current value; a magnetic pole position error estimating unit that estimates a magnetic pole position error estimate value using the estimated d-axis component, the estimated q-axis component, and the reference inductance value; A control device comprising:

2. The control device according to claim 1 , wherein the magnetic pole position error estimating unit estimates the magnetic pole position error estimated value by linear calculation of the estimated d-axis component and the estimated q-axis component.

3. The control device according to claim 1 , wherein the inductance reference unit provides a reference inductance value according to a d-axis component and a q-axis component of the detected current value.

4. The control device according to claim 1 , wherein the inductance reference unit provides a reference inductance value according to a d-axis component and a q-axis component of the current command value.

Images