JP2010166638A - Controller of rotating electrical machine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a controller of a rotating electrical machine which can markedly suppress the influence of ripples and can drive-control the rotating electrical machine to a desired torque/speed, at correcting of a position error due to position estimation that utilizes saliency. <P>SOLUTION: The controller is provided with a position estimating means 3 estimating a position of a rotor, based on the rotating electrical machine current flowing in the rotating electrical machine 1, which is detected by a current detecting means 2, and with a control means 5 for outputting a prescribed voltage command to a voltage applying means 6, based on the estimated estimation position. The control means 5 is provided with a current command computing part 7 for computing a current command, based on one command among a position command, a speed command and a torque command, a position estimation error computing part 23 for computing a position estimation error based on the current command, and a current command correcting part 9 for outputting a corrected current command, obtained by correcting the current command based on the position estimation error. <P>COPYRIGHT: (C)2010,JPO&INPIT

Description

  The present invention relates to a control device for a rotating electrical machine that performs rotation control by obtaining rotor position information without using a rotational position sensor in a rotating electrical machine such as an induction machine or a synchronous machine.

  In order to accurately control the rotating operation of the rotating electrical machine, information on the rotor position of the rotating electrical machine and information on the current flowing through the rotating electrical machine are required. Here, the rotor position information is conventionally obtained by attaching a rotation position sensor to the rotating electrical machine. However, the separate provision of the rotational position sensor is disadvantageous from the viewpoint of cost reduction, space saving, and improvement of reliability, and therefore, the rotational position sensor is required to be sensorless.

  As a control method for reducing the rotational position sensor in the rotating electrical machine, a method of estimating the rotor position of the rotating electrical machine mainly from an induced voltage of the rotating electrical machine, and a rotor position of the rotating electrical machine using saliency are used. There are two ways to estimate. Since the magnitude of the induced voltage used in the former method is proportional to the speed of the rotating electrical machine, the induced voltage decreases at the zero speed and low speed ranges, the S / N ratio deteriorates, and the rotor position of the rotating electrical machine Is difficult to estimate. On the other hand, in the latter method using the saliency, a position estimation signal for estimating the rotor position of the rotating electrical machine must be injected into the rotating electrical machine. However, the rotor of the rotating electrical machine is independent of the speed of the rotating electrical machine. Since the position can be estimated, a position sensorless control method using saliency is used at zero speed or low speed.

  However, in the position estimation method using saliency, the stator and rotor are magnetically saturated when a current is supplied to the rotating electrical machine, and the inductance distribution of the rotating electrical machine changes. An error (hereinafter referred to as a position estimation error) occurs. Therefore, if this position estimation error is not corrected, a desired torque and speed cannot be obtained, and the rotating electrical machine will step out in some cases.

  As a countermeasure, the q-axis current command and the value obtained by multiplying the q-axis current by a proportional constant are used to determine the position based on the fact that there is a proportional relationship between the q-axis current command and the position estimation error due to the mutual interference of the inductance. A technique has been proposed in which the estimated position is directly corrected using the position estimation error as a correction amount (see, for example, Patent Document 1 and Non-Patent Document 1 below).

JP 2002-291283 A

Z. Q. Zhu, Y. et al. Li, D.D. Howe, and M, Bingham: "Compensation for Rotor Position Estimation Error due to Cross-Coupling Magnetic Saturation in Signal Injection Based Sensorless Control of PM Brushless AC Motors", Proceediongs of 2007 IEEE Internationl Electric Machines and Drives Conference, Vol-1, pp . 208-213 (2007-5)

  In the position estimation method using saliency, in Patent Document 1 and Non-Patent Document 1 described above, a position estimation error is obtained as a correction amount from a value obtained by multiplying a q-axis current command by a proportional constant, and the estimated position is directly corrected. Therefore, if the pulsating component is superimposed on the q-axis current command, the pulsating component is included in the position estimation error. Therefore, if the estimated position is directly corrected with this position estimation error, the pulsating component is also applied to the estimated position. Will occur. As a result, by performing coordinate conversion of the d-axis voltage command and the q-axis voltage command into a three-phase voltage command based on this estimated position, this pulsation component is also superimposed on the drive voltage applied to the rotating electrical machine, resulting in torque. Pulsation and noise are generated, and in some cases, the pulsation component superimposed on the estimated position interferes with the pulsation component superimposed on the q-axis current command, resulting in a form in which the pulsation component is superimposed twice, which is An excessive current may flow.

  The present invention has been made in order to solve the above-described problem, and when performing position error correction accompanying position estimation using saliency, it is possible to suppress the influence of pulsation as much as possible, and to rotate. An object of the present invention is to provide a control device for a rotating electrical machine that can drive and control the electrical machine to a desired torque and speed.

  A control device for a rotating electrical machine according to the present invention controls rotation of the rotating electrical machine, and includes voltage applying means for applying a rotational driving voltage to the rotating electrical machine, and a rotating electrical machine current flowing through the rotating electrical machine. Current detecting means for detecting the position of the rotor, position estimating means for estimating the position of the rotor based on the rotating electrical machine current detected by the current detecting means, and voltage applying means based on the estimated position estimated by the position estimating means. And a control means for outputting a predetermined voltage command. The control means includes a current command calculation unit for calculating a current command based on any one of a position command, a speed command, and a torque command, and the current command. A position estimation error calculation unit for calculating a position estimation error based on the current command from the command calculation unit, and the current command based on the position estimation error calculated by the position estimation error calculation unit. And a current instruction correction unit which outputs the corrected correction current command, is characterized by calculating a voltage command for driving the rotary electric machine on the basis of the above correction current command and the estimated position.

  In the control apparatus for a rotating electrical machine according to the present invention, in the position estimation method using saliency, an inductance distribution is changed by passing a current through the rotating electrical machine, and a position estimation error caused by mutual interference of inductance due to magnetic saturation occurs. In this case, the estimated position is not directly corrected using the position estimation error as a correction amount as in the conventional case, but the current command is corrected according to the position estimation error. For this reason, even if the pulsation component is superimposed on the position estimation error as a result of the pulsation component being included in the current command, the estimated position where the d-axis voltage command and the q-axis voltage command are converted into a three-phase voltage command. Is not affected by pulsation. Therefore, when a drive voltage is applied to the rotating electrical machine based on the estimated position, the influence of the pulsation component is suppressed, and the generation of torque pulsation and noise is reduced. Moreover, since the pulsation component superimposed on the estimated position and the pulsation component superimposed on the q-axis current command do not interfere with each other as in the conventional case, the pulsation component is not superimposed twice. Thus, the rotary electric machine can be driven and controlled to a desired torque and speed.

It is a block diagram of the control apparatus of the rotary electric machine in Embodiment 1 of this invention. It is a block diagram which shows the detail of the position estimation means in the same control apparatus. 6 is an explanatory diagram for explaining an operation of the first embodiment. FIG. 6 is an explanatory diagram for explaining an operation of the first embodiment. FIG. FIG. 6 is a configuration diagram illustrating a modification of the control device for the rotating electrical machine according to the first embodiment. FIG. 6 is a configuration diagram illustrating another modification of the control device for the rotating electrical machine according to the first embodiment. FIG. 6 is a configuration diagram illustrating still another modification of the control device for the rotating electrical machine according to the first embodiment. It is a block diagram of the control apparatus of the rotary electric machine in Embodiment 2 of this invention.

Embodiment 1 FIG.
FIG. 1 is a configuration diagram of a control device for a rotating electrical machine according to Embodiment 1 of the present invention, and FIG. 2 is a configuration diagram showing details of position estimation means of the control device.

  In the first embodiment, the rotating electrical machine 1 is, for example, an embedded magnet type synchronous machine, and the rotating electrical machine 1 is supplied with voltage commands Vu *, Vv *, Vw * output from a control means 5 described later. A voltage applying means 6 for applying a predetermined control voltage for rotational driving is connected based on (hereinafter, symbol * means each command value). This voltage application means 6 is comprised by power converters, such as an inverter, for example.

  Further, the control device according to the first embodiment is rotated by a current detecting means 2 for detecting a rotating electrical machine current flowing between the voltage applying means 6 and the rotating electrical machine 1 and a rotating electrical machine current detected by the current detecting means 2. A position estimating unit 3 that detects an inductance change of the electric machine 1 and estimates a position of the rotor of the rotating electric machine 1 based on the detected inductance change, and a voltage application based on the estimated position estimated by the position estimating unit 3. The means 6 comprises control means 5 for outputting predetermined voltage commands Vu *, Vv *, Vw *.

  The current detection means 2 comprises a current transformer, for example. In this example, the U-phase rotating electrical machine current iu and the W-phase rotating electrical machine current iw are detected from the power line connecting the rotating electrical machine 1 and the voltage applying means 6. To do. However, in addition to this, the rotating electrical machine current for any two phases of the U phase, the V phase, and the W phase may be detected. Alternatively, a method of detecting all the rotating electrical machine currents for the three phases of the U phase, the V phase, and the W phase may be used. Alternatively, as the current detection means 2, it is possible to detect a DC bus current that is an input of the voltage application means 6 and to calculate the rotating electrical machine current from the DC bus current.

  The position estimation means 3 estimates the position of the rotor of the rotating electrical machine 1 using the property (saliency) that the inductance of the rotating electrical machine 1 changes depending on the position of the rotor. Therefore, in the first embodiment, in the adder 14 of the control means 5, the high frequency voltages Vuh, Vvh, Vwh as the three position estimation signals output from the high frequency voltage generator 13 are converted into the dq / three-phase converter 12. Is superimposed on the voltage commands Vu *, Vv *, and Vw * output from. If the rotating electrical machine current is detected by the current detection means 2, the rotating electrical machine current includes currents having the same frequency components as the high frequency voltages Vuh, Vvh, Vwh output from the high frequency voltage generator 13 (hereinafter referred to as high frequency current). Therefore, the position estimation means 3 extracts the high-frequency current, estimates the position of the rotor using the fact that the amplitude of the extracted high-frequency current depends on the position of the rotor of the rotating electrical machine 1, The estimated position θh is output.

Hereinafter, the configuration of the position estimation unit 3 and the contents of the position estimation process will be described in detail.
As shown in FIG. 2, the position estimation means 3 includes a three-phase / two-phase converter 15, two current amplitude extractors 16, two multipliers 17, a subtractor 18, and a position calculator 19.

Here, the three-phase / two-phase converter 15 converts the rotating electrical machine currents iu, iw detected by the current detection means 2 into two-phase currents iαs, iβs. Thus, the calculation process required for the subsequent position estimation process can be simplified by converting into a two-phase current. Then, each current amplitude extractor 16 applies high-frequency voltages Vuh, Vvh, Vwh for position estimation from the two-phase currents iαs, iβs that are the outputs of the three-phase / two-phase converter 15, thereby rotating the electric machine 1. Are extracted by using Fourier transform or the like. Subsequently, each of the multipliers 17a and 17b squares Iαs and Iβs that are the outputs of the current amplitude extractors 16a and 16b, respectively, and outputs (Iαs) ² and (Iβs) ² . The subtracter 18 subtracts (Iαs) ² from (Iβs) ² and outputs ΔIαβ. The position calculator 19 calculates the estimated position θh from ΔIαβ that is the output of the subtractor 18.

  The rotor position θh obtained in this way is not a value detected directly using a sensor or the like, but is an estimated value obtained by calculation, but this method also estimates the correct rotor position. be able to. Hereinafter, the content of the position estimation process will be described more specifically.

  When the rotating electrical machine 1 is, for example, an embedded magnet type synchronous machine, the voltage equation at fixed orthogonal coordinates (α-β axes) can be expressed as the following equation (1).

  Assuming that the rotating electrical machine is stopped or operating at low speed, ω = 0, and if the differential operator P is replaced with the Laplace operator s, the currents iαs and iβs in the fixed orthogonal coordinates are expressed by the following equation (2). .

  Now, when the high frequency voltages Vuh, Vvh, Vwh having an angular frequency ωh sufficiently higher than the angular frequency of the voltage commands Vu *, Vv *, Vw * for driving the rotating electrical machine 1 from the high frequency voltage generator 13 are applied, R << Lα · ωh and R << Lβ · ωh hold (when s = jωh (where j is an imaginary unit)), and when the influence of the stator resistance R is ignored, the above equation (2) is expressed by the following (3) It becomes an expression.

  Further, the high frequency voltages Vuh, Vvh, Vwh applied from the high frequency voltage generator 13 can be expressed by the following equation (4) in fixed orthogonal coordinates.

  The high frequency voltage in the fixed orthogonal coordinates of the equation (4) is expressed by the following equation (5).

  Therefore, when the value of the high-frequency voltage in the equation (5) is substituted into [Vαs Vβs] in the fixed orthogonal coordinate system in the equation (3) and s = jωh (j is an imaginary unit), the following equation (6) is obtained. .

As shown in equation (6), it can be seen that the rotor position information θ (= rotor position θh) is included in the amplitudes of the currents iαs and iβs in the fixed orthogonal coordinates. Therefore, the amplitudes Iαs and Iβs are extracted from the high-frequency currents iαs and iβs generated in the rotating electrical machine 1 by applying the position estimation high-frequency voltages Vuh, Vvh, and Vwh by the current amplitude extractors 15. Then, based on the extracted amplitudes Iαs and Iβs, a term including only the rotor position information θ (rotor position θh) is extracted by performing an operation as shown in the following equation (7). In order to realize this calculation, each multiplier 16 that squares the amplitudes Iαs and Iβs respectively, and (Iβs) ² that is the output of each multiplier 16 is subtracted from (Iαs) ² to obtain only the information on the rotor position θ. A subtractor 18 that outputs ΔIαβ is used.

  The position calculator 19 extracts only cos 2θ by dividing ΔIαβ in the above equation (7) by the following equation (8). Then, θ (rotor position θh) is calculated by calculating the inverse cosine of cos 2θ.

  The calculation of the rotor position θh is not an inverse cosine calculation, but a table in which the value of cos 2θ is stored is prepared, and the rotor position θh is obtained based on the value of cos 2θ stored in the storage device, or Taylor expansion Alternatively, the rotor position θh may be obtained by performing an approximate calculation such as equation (9) based on Taylor expansion.

  The position estimating means 3 is not limited to the method for obtaining the rotor position θh as described above, but may be one for obtaining the rotor position θh based on the rotating electrical machine current detected by the current detecting means 2. For example, a method other than the above may be used.

Next, the configuration and control operation of the control means 5 will be described.
In the first embodiment, the control means 5 includes a current command calculation unit 7, a position estimation error calculation unit 23, a current command correction unit 9, a three-phase / dq converter 10, a voltage command calculation unit 11, a dq / three-phase conversion. And a high frequency voltage generator 13 and an adder 14.

  The current command calculation unit 7 outputs current commands id * and iq * on dq coordinates based on a torque command Tref given from the outside. That is, when the torque command Tref is input from the outside, the current command calculation unit 7 confirms that the relationship between the output torque T of the rotating electrical machine 1 and the current commands id and iq on the dq coordinate has the following equation (10). Use to determine id * and iq *. For example, in the case of controlling id * = 0, iq * is determined as in equation (11) using equation (10). As will be described later, it is also possible to obtain the current commands id * and iq * based on the position command or the speed command in addition to the torque command Tref.

Ｐｍ：回転電機の極対数
φｍ：磁石磁束の大きさ

Pm: Number of pole pairs of rotating electric machine φm: Size of magnetic flux

  By the way, as described above, the position estimating unit 3 extracts the high frequency current included in the rotating electrical machine current detected by the current detecting unit 2, and the amplitude of the extracted high frequency current depends on the position of the rotating electrical machine 1. That is, the position of the rotor of the rotating electrical machine 1 is estimated using the fact that the current vector locus of the high-frequency current flowing through the rotating electrical machine 1 becomes an ellipse as a result of the inductance changing depending on the position of the rotor, and the estimated position Output as θh.

  Here, when there is no load in the rotating electrical machine 1 and the d-axis current id = 0 and the q-axis current iq = 0, the current vector locus of the high-frequency current is the long axis as the d-axis as shown in FIG. , An ellipse in the dq axis coordinate system with the short axis as the q axis (FIG. 3 shows a case where Lq> Ld). However, when the current command id *, iq * is actually applied to drive the rotating electrical machine 1 and the current (d-axis current id ≠ 0, q-axis current iq ≠ 0) flows to the rotating electrical machine 1, FIG. ), The current vector locus of the high-frequency current is an ellipse in the dh-qh axis coordinate system in which the major axis is the dh axis and the minor axis is the qh axis, and the d axis and the dh axis, and the q axis and the qh axis. The directions will not match each other. That is, a position estimation error ΔθL corresponding to the magnitudes of the current commands id * and iq * occurs between the estimated position θh output from the position estimating means 3 and the actual position of the rotating electrical machine 1.

  In other words, the d-axis current command id * and the q-axis current command iq * output from the current command calculation unit 7 are, as shown in FIG. It is assumed that the coordinate system is a dq axis coordinate system. However, the position estimation error ΔθL is generated by giving the current commands id * and iq * according to the load of the rotating electrical machine 1, and the current vector locus of the high-frequency current has a long axis as shown in FIG. When an ellipse in the dh-qh axis coordinate system with the dh axis and the short axis as the qh axis is used, the current command id * for the d axis should be given to the dh axis and the current to the q axis. What should give command iq * is given to the qh axis. As a result, the rotating electrical machine 1 cannot be controlled to a desired torque or speed, and in some cases, a problem occurs that the step-out occurs. Therefore, when the position estimation error ΔθL occurs due to the current commands id * and iq * to be given according to the load of the rotating electrical machine 1, the current commands id * and iq * are corrected according to the magnitude of the position estimation error ΔθL. The current commands id * and iq * need to be given to the original dq axis system shown in FIG.

  Therefore, in the first embodiment, the position estimation error calculation unit 23 and the current command correction unit 9 are provided to correct the current commands id * and iq * according to the magnitude of the position estimation error ΔθL. .

Next, the details of the correction will be described in more detail.
First, when the positional estimation error ΔθL is proportional to the q-axis current command iq *, the position estimation error calculation unit 23 sets the proportional coefficient K1 to iq * as shown in the following equation (12). The multiplied product is output as ΔθL. However, the present invention is not limited to this, and when a proportional relationship is established between the position estimation error ΔθL and the d-axis current command id *, a value obtained by multiplying id * by a proportional coefficient K2 as shown in equation (13). You may output as (DELTA) (theta) L. If the estimated position error ΔθL is proportional to both id * and iq *, ΔθL may be obtained and output as shown in equation (14). Thus, when the proportional relationship is established between the position estimation error ΔθL and the q-axis current command iq * or the d-axis current command id *, the position estimation error ΔθL can be easily calculated.

  In addition, the position estimation error calculation unit 23 sets ΔθL to iq * if the proportional relationship does not hold between the estimated position error ΔθL and the q-axis current command iq * or the d-axis current command id *. The estimated position error Δθ generated according to the value of id * or iq * is stored in advance as a table, and ΔθL is read from the table stored according to the value of id * or iq *. May be output. If the table is formed in advance in this way, the position estimation error ΔθL can be obtained more easily based on the q-axis current command iq * and the d-axis current command id *.

  Next, the current command correction unit 9 performs the calculation shown in the following equation (15) based on the position estimation error ΔθL that is the output of the position estimation error calculation unit 23 to perform the d-axis current command id * and the q-axis current. The command iq * is corrected, and after correction, it is output as corrected current commands id_c * and iq_c *.

  The operation principle will be described with reference to FIG. FIG. 4 illustrates a case where id * = 0 is given as the d-axis current command and iq * is given as the q-axis current command in order to facilitate understanding and simplify the explanation.

  When there is an estimated position error ΔθL, as shown in FIG. 4A, the coordinate system based on the estimated position θh output by the position estimating means 3 is a dh-qh axis coordinate in which the long axis is the dh axis and the short axis is the qh axis. Become a system. On the other hand, the d-axis current command id * and the q-axis current command iq * output from the current command calculation unit 7 are a dq-axis coordinate system in which the d-axis is the long axis and the q-axis is the short axis. Therefore, if no correction is made to the q-axis current command iq *, the q-axis current command iq * will flow on the qh axis even though it should originally flow on the q-axis.

  In order to eliminate this inconvenience, the current command correction unit 9 uses ΔθL that is the output of the position estimation error calculation unit 23 to obtain a d-axis correction current command that is a current command on the dh-axis as shown in Equation (15). Each of id_c * and q-axis corrected current command iq_c *, which is a current command on the qh axis, is calculated and output. In this way, for example, when the d-axis current command is id * = 0 and the q-axis current command is iq *, as shown in FIG. 4B, the d-axis correction current that is the current command on the dh-axis is shown. A combined vector of the command id_c * and the q-axis corrected current command iq_c * which is a current command on the qh axis is iq * on the q-axis. That is, the q-axis current command iq * can be sent to the q-axis of the original dq-axis coordinate system.

  In this way, the current command correction unit 9 calculates and outputs the d-axis correction current command id_c * and the q-axis correction current command iq_c *, which are combined vectors, based on the equation (15).

  On the other hand, the three-phase / dq converter 10 uses the estimated position θh, which is the output of the position estimating means 3, to convert the rotating electrical machine current (static coordinates) detected by the current detecting means 2 into a current in the dq axis coordinate system. Convert to some id, iq.

  The voltage command calculation unit 11 is a difference between id_c * and iq_c * that are outputs of the current command correction unit 9 and id and iq that are outputs of the three-phase / dq converter (id_c * -id and iq_c-iq). Is calculated and output by calculating the d-axis voltage command Vd * and the q-axis voltage command Vq *.

  The dq / three-phase converter 12 uses the estimated position θh, which is the output of the position estimating means 3, to convert the Vd *, Vq * (dq coordinate), which is the output of the voltage command calculation unit 11, into the three-phase voltage command Vu *, Convert to Vv *, Vw * (stationary coordinates).

  The high-frequency voltage generator 13 is used to estimate the estimated position θh from the position estimating means 3, and is higher than the frequencies of Vu *, Vv *, and Vw * that are the outputs of the dq / three-phase converter 12. The high frequency voltages Vuh, Vvh, Vwh of the frequency are output.

  The adder 14 adds three-phase voltage commands Vu *, Vv *, Vw * which are outputs of the dq / three-phase converter 12 and Vu, Vvh, Vwh which are outputs of the high-frequency voltage generator 13, respectively. Vvs and Vws are output to the voltage applying means 6.

  The voltage applying means 6 applies a voltage to the rotating electrical machine by a power conversion device such as an inverter based on the voltage commands Vus, Vvs, Vvs that are outputs of the control means 5.

  As described above, in the first embodiment, the correct d-axis and q-axis of the rotating electrical machine 1 can be obtained even if the position estimation error ΔθL occurs by giving the current commands id_c * and iq_c * according to the load of the rotating electrical machine 1. The d-axis current command id * and the q-axis current command iq * can be flowed upward. Therefore, the rotating electrical machine 1 can be controlled to a desired torque and speed, and the rotating electrical machine 1 can be driven without stepping out.

  Further, instead of directly correcting the estimated position θh using the position estimation error ΔθL as a correction amount as in the prior art, the current commands id * and iq * are corrected according to the position estimation error ΔθL. Even if pulsation components are included in the current commands id * and iq * calculated by the calculation unit 7, and the pulsation component is superimposed on the position estimation error ΔθL, the estimated position θh is affected by the pulsation. I do not receive it. Therefore, when a drive voltage is applied to the rotating electrical machine 1 based on the estimated position θh, the influence of the pulsation component is suppressed, the generation of torque pulsation and noise is reduced, and no excessive current flows through the rotating electrical machine 1. The rotary electric machine 1 can be driven and controlled to a desired torque and speed.

  In the first embodiment, the current command calculation unit 7 is configured to obtain the current commands id * and iq * on the dq coordinate based on the torque command Tref given from the outside. For example, when a speed command ωref is given from the outside, as shown in FIG. 5, a speed estimator 18 is provided in the control means 5, and the speed estimator 18 sets an estimated position θh that is an output of the position estimating means 3. Based on this, an estimated speed ωh is obtained and output to the current command calculation unit 7, and the current command calculation unit 7 obtains a deviation Δω between the estimated speed ωh and the speed command ωref and performs proportional integral control on the deviation Δω. It is also possible to determine the current commands id * and iq *.

  Further, when the position command θref is given from the outside, as shown in FIG. 6, a speed estimator 18 is provided in the control means 5, and the speed estimator 18 sets an estimated position θh that is an output of the position estimation means 3. Based on this, the estimated speed ωh is obtained and output to the current command calculation unit 7, and the current command calculation unit 7 obtains a deviation Δθ between the estimated position θh and the position command θref which is the output of the position estimation means 3, and this deviation The speed command ωref is calculated by proportionally integrating Δθ, and then a deviation Δω between the speed command ωref and the estimated speed ωh given from the speed estimator 18 is obtained, and the current is obtained by proportionally integrating the deviation Δω. It is also possible to determine the commands id * and iq *.

  In the first embodiment, the position estimation error calculation unit 23 calculates the estimated position error ΔθL based on the d-axis current command id * and the q-axis current command iq *. Instead, For example, as shown in FIG. 7, it is also possible to obtain the estimated position error ΔθL using the d-axis current id and the q-axis current iq that are the outputs of the three-phase / dq converter 10. In FIG. 7, the torque command is input to the current command calculation unit 7. However, as described in the first embodiment, the position command or the speed command may be input by providing the speed estimator 18.

Embodiment 2. FIG.
FIG. 8 is a block diagram showing a control device for a rotating electrical machine according to the second embodiment of the present invention. Components corresponding to or corresponding to those of the first embodiment shown in FIG.

  In the first embodiment, the position estimation error calculator 23 is proportional to the d-axis current command id * and the q-axis current command iq * as shown in, for example, the equations (12) to (14) described above. Is multiplied as the estimated position error ΔθL.

  On the other hand, in the second embodiment, as shown in FIG. 8, the position estimation error calculation unit 23 calculates the estimated position error ΔθL based on the rotating electrical machine currents iu and iw detected by the current detection means 2. It is what I did. Details of the position estimation error calculation unit 23 will be described below.

  As described in the first embodiment, when the current for driving the rotating electrical machine 1 flows to the rotating electrical machine 1, an estimated position error ΔθL is generated. Since this estimated position error ΔθL is generated in accordance with the current flowing through the rotating electrical machine 1, the position estimating error calculating unit 23 performs the following (16) based on the rotating electrical machine currents iu and iw detected by the current detecting means 2. The rotating electrical machine current effective value Ieff or the rotating electrical machine current peak value Iamp is calculated as shown in the equation or (17). Note that iv in the equations (16) and (17) can be obtained by utilizing the relationship of the equation (18) with the three-phase rotating electrical machine current.

  Next, as shown in the equation (19) or (20), the calculated Ieff or Iamp multiplied by the proportional coefficient K3 or K4 is output as the position estimation error ΔθL. As described above, when the position estimation error ΔθL has a certain relationship with the rotating electrical machine currents iu and iw, the position estimation error ΔθL can be easily calculated.

When the estimated positional error ΔθL and the rotary electric machine current effective value Ieff and the rotary electric machine current peak value Iamp do not have a proportional relationship as shown in the formulas (19) and (20) and are different functions. The position estimation error calculator 23 stores ΔθL as a function of Ieff or Iamp, or stores the estimated position error Δθ generated according to the value of Ieff or Iamp in advance as a table, and stores it according to Ieff or Iamp. Alternatively, Δθ may be read from the table and output. If a table is prepared in advance as described above, the position estimation error ΔθL can be obtained more easily. Here, the rotating electrical machine currents iu and iw related to the U phase and the W phase are used. However, the rotating electrical machine currents for any two phases of the U phase, the V phase, and the W phase may be detected. Moreover, the method of detecting all the rotating electrical machine currents for three phases of the U phase current, the V phase current, and the W phase current may be used. In FIG. 8, the torque command is input to the current command calculation unit 7. However, as described in the first embodiment, the position command or the speed command may be input by providing the speed estimator 18.
Since other components are the same as those in the first embodiment, detailed description thereof is omitted here.

  As described above, in the second embodiment, as in the case of the first embodiment, even if the position estimation error ΔθL is generated by giving the current commands id_c * and iq_c * according to the load of the rotating electrical machine 1, The d-axis current command id * and the q-axis current command iq * can flow on the correct d-axis and q-axis of the electric machine 1. Therefore, the rotating electrical machine 1 can be controlled to a desired torque and speed, and the rotating electrical machine 1 can be driven without stepping out. In addition, since the estimated position θh is not directly corrected by the estimated position error ΔθL, the current commands id * and iq * are corrected. Therefore, even if the estimated position error ΔθL is superimposed on the estimated position error ΔθL, the estimated position θh is not affected by the pulsation of the estimated position error ΔθL. For this reason, it becomes possible to control to a desired torque and speed without stepping out the rotating electrical machine 1. Further, generation of torque pulsation and noise can be reduced, and excessive current can be prevented from flowing through the rotating electrical machine 1.

1 rotating electrical machine, 2 current detection means, 3 position estimation means, 5 control means,
6 voltage application means, 7 current command calculation unit, 9 current command correction unit,
10 three-phase / dq converter, 11 voltage command calculation unit, 12 dq / three-phase converter,
13 high frequency voltage generator, 14 adder, 23 position estimation error calculator.

Claims (6)

Rotation control of the rotating electrical machine, the voltage applying means for applying a voltage for rotational drive to the rotating electrical machine, current detection means for detecting the rotating electrical machine current flowing through the rotating electrical machine, and the current detection Position estimating means for estimating the position of the rotor based on the rotating electrical machine current detected by the means, and control means for outputting a predetermined voltage command to the voltage applying means based on the estimated position estimated by the position estimating means; With
The control means includes a current command calculation unit that calculates a current command based on any one of a position command, a speed command, and a torque command, and a position estimation error based on the current command from the current command calculation unit. And a current command correction unit that outputs a corrected current command obtained by correcting the current command based on the position estimation error calculated by the position estimation error calculation unit. A control device for a rotating electrical machine that calculates a voltage command for driving the rotating electrical machine based on a current command and the estimated position. 2. The position estimation error calculation unit outputs a value obtained by multiplying a magnitude of the current command calculated by the current command calculation unit by a proportional constant as a position estimation error. The control apparatus of the rotary electric machine described. 2. The rotation according to claim 1, wherein the position estimation error calculation unit outputs a position estimation error stored in advance according to the magnitude of the current command calculated by the current command calculation unit. Electric control device. The control apparatus for a rotating electrical machine according to claim 1, further comprising a position estimation error calculation unit that calculates a position estimation error based on the rotating electrical machine current detected by the current detection means, instead of the position estimation error calculation unit. A control device for a rotating electrical machine. The position estimation error calculation unit outputs, as a position estimation error, a value obtained by multiplying the amplitude of the rotating electrical machine current detected by the current detection means by a proportionality constant. 4. A control device for a rotating electrical machine according to 4. 5. The position estimation error calculation unit outputs a position estimation error stored in advance according to the amplitude of the rotating electrical machine current detected by the current detection means. Rotating electrical machine control device.

Images