JP5512054B1 - AC rotating machine control device - Google Patents

Abstract

  The AC rotating machine control device uses AC rotating machine position information to generate an AC rotating machine control means for creating a voltage command vector so that the detected current vector matches the current command vector, and based on the voltage command vector, According to voltage application means for applying a voltage to the rotating machine, secondary magnetic flux vector calculating means for calculating a secondary magnetic flux vector in the AC rotating machine, the voltage command vector, the detected current vector, and the secondary magnetic flux vector Adaptive observation means for obtaining and outputting an estimated rotational position and an estimated speed, wherein the adaptive observation means can switch whether or not the secondary magnetic flux vector is used with a switching gain, and the adaptive observation means Compensating means for compensating for the fluctuation of the estimated speed that occurs when the switching gain is switched.

Description

  The present invention relates to a control device for an AC rotating machine.

  Known as a method for handling the voltage and current of an AC rotating machine on the rotating biaxial coordinates (dq axes) and controlling the torque generated by an AC rotating machine such as a permanent magnet synchronous rotating machine with high response and high accuracy. There is a vector control that is When performing vector control of a permanent magnet synchronous rotating machine, it is necessary to attach a position sensor to the AC rotating machine in order to use rotor rotational position information when performing voltage and current coordinate conversion. However, attaching the position sensor has disadvantages such as an increase in cost, an increase in dimensions of the AC rotating machine, and a decrease in maintainability. Therefore, in order to eliminate this disadvantage, a method of performing vector control without a position sensor by estimating a rotational position in a control device has been developed.

Patent Document 1 discloses a control system for a synchronous motor, the adaptive observer, the d-axis voltage command on the rotation two-axis coordinate (d-q-axis), the q-axis voltage command, based on the d-axis current and q-axis current It is described that the angular frequency of the rotor is obtained and output, and the integrator obtains and outputs the rotational position of the rotor by integrating the angular frequency of the rotor. Thus, according to Patent Document 1, since the configuration of the adaptive observer on a rotating two-axis coordinates, can the frequency components of the voltage input to the adaptive observer even when operating at high rotational speeds in DC, inexpensive Even when a computer is used, the synchronous motor can be controlled at a high rotational speed.

  In Patent Document 2, in an AC rotating machine control apparatus, adaptive observation means uses an estimated magnetic flux phase, an estimated current vector, an estimated magnetic flux vector, and an estimated speed based on an amplified deviation vector, a current deviation vector, and a voltage command vector. Is output. At this time, the rotational position detection means detects the rotational position of the AC rotating machine, the magnetic flux vector detection means detects the magnetic flux vector from the detected rotational position, and outputs the detected magnetic flux vector to the adaptive observation means. The adaptive observation means gives a predetermined magnitude to the amplification gain that is multiplied by the detected magnetic flux vector and the magnetic flux deviation vector of the estimated magnetic flux vector in the speed range where the estimated speed is small, and multiplies the magnetic flux deviation vector in the other speed ranges. Set the amplification gain to zero. Thus, according to Patent Document 1, in the low speed range, it is possible to reliably control the AC rotating machine by calculating the estimated magnetic flux phase via the magnetic flux deviation vector that is reliably generated, and in the high speed range, It is said that even if the position detection accuracy of the rotational position detecting means is reduced, the estimated magnetic flux phase is calculated without using the detected magnetic flux vector, so that the AC rotating machine can be controlled stably.

International Publication No. 2002/091558 International Publication No. 2010/109528

  Patent Document 1 describes a method for realizing position sensorless control by estimating the rotational position and speed of a permanent magnet synchronous rotating machine with an adaptive observer operating on rotating biaxial coordinates. However, since there is no current deviation that is an estimation error of the adaptive observer at zero speed, it may be difficult to estimate the rotational position. In addition, stability and drive performance may be reduced due to the influence of voltage error or constant error in the low speed range.

  In Patent Document 2, a rotating machine secondary magnetic flux vector created from position information estimated by using the saliency of second rotating position detecting means, for example, a rotating position detector or a permanent magnet synchronous rotating machine directly attached to an AC rotating machine. Describes a method for operating the adaptive observer stably in the low-speed range including zero speed. In this method, in the middle / high speed range where the adaptive observer can estimate the rotational position independently, the amplification gain multiplied by the magnetic flux deviation vector is set to 0, so that the second rotational position detecting means is not used. It is possible to switch to operating only the adaptive observer.

  The invention described in Patent Document 2 can solve the problem of Patent Document 1, but since the rotating machine secondary magnetic flux vector created by the second rotational position detecting means is added to the input of the speed estimator, the amplification gain is increased. When it is set to 0 and the adaptive observer is switched to the single operation, the term including the rotating machine secondary magnetic flux vector is instantaneously lost from the input of the speed estimator, and the discontinuity of the estimated speed may increase.

  This invention is made | formed in view of the above, Comprising: It aims at obtaining the control apparatus of the alternating current rotating machine which can suppress the discontinuity of estimated speed.

  In order to solve the above-described problems and achieve the object, an AC rotating machine control device according to one aspect of the present invention uses an AC rotating machine position information to match a detected current vector with a current command vector. AC rotating machine control means for generating a voltage command vector, voltage applying means for applying a voltage to the AC rotating machine based on the voltage command vector, and a secondary magnetic flux vector for calculating a secondary magnetic flux vector in the AC rotating machine Computing means; and adaptive observation means for obtaining and outputting an estimated rotational position and an estimated speed according to the voltage command vector, the detected current vector, and the secondary magnetic flux vector, and the adaptive observation means, Whether or not a secondary magnetic flux vector is used can be switched by a switching gain, and the adaptive observation means can calculate a variation in the estimated speed that occurs when the switching gain is switched. It characterized in that it has a compensating means for amortization.

  According to the present invention, the compensation means compensates for the fluctuation in the estimated speed that occurs when the switching gain is switched. For example, the integral addition amount calculator calculates a compensation amount so as to compensate for a fluctuation in estimated speed that occurs when the switching gain is switched, generates an addition amount using the compensation amount, and supplies it to the speed estimator. To do. As a result, the speed estimator can perform proportional integral control and add the added amount to the integral term in the result of the proportional integral control to obtain the estimated speed. That is, since the estimated speed can be obtained while compensating for the fluctuation of the estimated speed that occurs when the switching gain is switched, the discontinuity of the estimated speed can be suppressed.

FIG. 1 is a diagram illustrating a configuration of a control device for an AC rotating machine according to a first embodiment. FIG. 2 is a diagram showing a configuration of the adaptive observer in the first embodiment. FIG. 3 is a vector diagram showing the secondary magnetic flux vector in the first embodiment. FIG. 4 is a diagram showing the relationship between the switching gain Kw and the estimated speed ωr ^ in the first embodiment. FIG. 5 is a diagram showing the operation in the first embodiment with time on the horizontal axis. FIG. 6 is a diagram showing a configuration of the adaptive observer in the second embodiment. FIG. 7 is a diagram showing the addition amount α ′ when the attenuator is a first-order lag element in the second embodiment. FIG. 8 is a diagram showing the addition amount α ′ when the attenuator is linear attenuation in the second embodiment. FIG. 9 is a diagram showing a configuration of an adaptive observer in the third embodiment. FIG. 10 is a diagram illustrating the operation of Kwfil in the third embodiment.

  Hereinafter, embodiments of a control device for an AC rotating machine according to the present invention will be described in detail with reference to the drawings. Note that the present invention is not limited to the embodiments.

Embodiment 1 FIG.
A control device 100 for an AC rotating machine according to a first embodiment will be described.

  The control device 100 is a control device that drives an AC rotating machine (for example, a permanent magnet synchronous rotating machine) M at a variable speed. For example, the control device 100 uses the estimated rotation position by the adaptive observation means (for example, the adaptive observation means 33) and the detected rotation position of the AC rotating machine M detected by a different second means (for example, the rotation position detection means 34). It has the function of performing vector control in combination.

  Specifically, the control device 100 has the configuration shown in FIG. FIG. 1 is a diagram illustrating an overall configuration of a control device 100 for an AC rotating machine. In the following description, the direction of the rotating machine secondary magnetic flux vector estimated by the adaptive observer 11 is defined as the d axis, and the direction orthogonal thereto is defined as the q axis. Further, for example, an α axis-β axis coordinate system is considered as an arbitrary fixed two-axis orthogonal coordinate system, for example, the α axis is set to 0 [rad], and the phase is an angle from the α axis to the d axis. At this time, the u axis of the fixed three-axis orthogonal coordinate system (u-axis-v-axis-w-axis coordinate system) may coincide with the α-axis.

  The control device 100 includes an AC rotating machine control means 31, a voltage application means 5, a current detection means 6, a rotational position detection means 34, a secondary magnetic flux vector calculation means 32, and an adaptive observation means 33.

  The AC rotating machine control means 31 uses the AC rotating machine position information to create a voltage command vector so that the detected current vector matches the current command vector. For example, the AC rotating machine control means 31 includes an id command calculator 1, a speed controller 2, a 3-phase / 2-phase converter 8, a current controller 3, and a 2-phase / 3-phase converter 4.

  The id command calculator 1 creates a d-axis current command id * by id = 0 control or the like that always controls id to 0. The id command calculator 1 supplies the created d-axis current command id * to the current controller 3.

  The speed controller 2 receives the speed command ω * from the outside (for example, a host controller (not shown)) and receives the estimated speed ωr ^ from the adaptive observation means 33. The speed controller 2 creates a q-axis current command iq * so that the estimated speed ωr ^ follows the speed command ω *. The speed controller 2 supplies the created q-axis current command iq * to the current controller 3.

  The three-phase / two-phase converter 8 receives the detected currents iu, iv, iw from the current detection means 6 and receives the estimated rotational position θ1 from the adaptive observation means 33 as AC rotating machine position information. The two-phase / three-phase converter 4 converts the detected current vector (iu, iv, iw) of the u-axis-v-axis-w-axis coordinate system (fixed coordinate system) based on the estimated rotational position θ1 to the d-axis-q-axis. Coordinates are converted to a detected current vector (id, iq) in the coordinate system (rotating coordinate system). The three-phase / two-phase converter 8 supplies the converted detected current vector (id, iq) to the current controller 3 and the adaptive observation means 33.

  The current controller 3 receives the d-axis current command id * from the id command computing unit 1, receives the q-axis current command iq * from the speed controller 2, and receives the d-axis detection current id and the q-axis detection current iq in three phases / Received from the two-phase converter 8. The current controller 3 creates the d-axis voltage command vd * so that the d-axis detection current id follows the d-axis current command id *, and causes the q-axis detection current iq to follow the q-axis current command iq *. Q-axis voltage command vq * is created.

  That is, the current controller 3 creates the voltage command vector (vd *, vq *) so that the detected current vector (id, iq) matches the current command vector (id *, iq *). The current controller 3 outputs the created voltage command vector (vd *, vq *) to the 2-phase / 3-phase converter 4 and the adaptive observation means 33.

  The two-phase / three-phase converter 4 receives the voltage command vector (vd *, vq *) from the current controller 3 and receives the estimated rotational position θ1 from the adaptive observation means 33 as AC rotating machine position information. The 2-phase / 3-phase converter 4 converts the voltage command vector (vd *, vq *) of the d-axis-q-axis coordinate system (rotation coordinate system) based on the estimated rotation position θ1 into the u-axis-v-axis-w-axis. Coordinates are converted to voltage command vectors (vu *, vv *, vw *) in the coordinate system (fixed coordinate system). The two-phase / three-phase converter 4 supplies the converted voltage command vector (vu *, vv *, vw *) to the voltage applying means 5.

  The voltage application means 5 receives voltage command vectors (vu *, vv *, vw *) from the two-phase / three-phase converter 4. The voltage applying means 5 applies a voltage to the AC rotating machine M based on voltage command vectors (vu *, vv *, vw *). In response to this, power is supplied from the voltage application means 5 to the AC rotating machine M, and the AC rotating machine M is driven.

  The current detection means 6 detects currents iu, iv, iw flowing through the AC rotating machine. The current detection means 6 includes, for example, a plurality of current detectors (for example, a plurality of current transformers) 6u to 6w, and the currents iu, iv, iw flowing through the AC rotating machine using the plurality of current detectors 6u to 6w. Is detected. The current detection means 6 supplies the detected currents iu, iv, iw to the three-phase / two-phase converter 8.

  The rotational position detector 34 detects the rotational position θr of the AC rotating machine. The rotational position detector 34 includes, for example, a rotational position detector (for example, an encoder) 9 and detects the rotational position θr of the AC rotating machine using the rotational position detector 9. The rotational position detector 34 supplies the detected rotational position θr to the secondary magnetic flux vector calculator 32.

  The secondary magnetic flux vector calculating means 32 calculates the secondary magnetic flux vector in the AC rotating machine M according to the detected rotational position θr and the estimated rotational position θ1, for example, according to the deviation between the detected rotational position θr and the estimated rotational position θ1. . For example, the secondary magnetic flux vector calculation means 32 includes a subtractor 22 and a secondary magnetic flux calculator 10.

  The subtractor 22 receives the detected rotation position θr from the rotation position detection unit 34 and receives the estimated rotation position θ1 from the adaptive observation unit 33. The subtractor 22 subtracts the estimated rotational position θ1 from the detected rotational position θr to obtain a position deviation Δθ and supplies it to the secondary magnetic flux calculator 10.

  Secondary magnetic flux calculator 10 receives position deviation Δθ from subtractor 22. The secondary magnetic flux calculator 10 creates a dq-axis secondary magnetic flux vector φrL ^ from the position deviation Δθ. The secondary magnetic flux calculator 10 supplies the dq axis secondary magnetic flux vector φrL ^ to the adaptive observation means 33.

  The adaptive observation means 33 determines the estimated rotational position θ1 and the estimated speed ωr ^ according to the voltage command vector (vd *, vq *), the detected current vector (id, iq), and the dq-axis secondary magnetic flux vector φrL ^. Find and output. For example, the adaptive observation unit 33 includes the adaptive observer 11 and the integrator 7.

  The adaptive observer 11 receives the voltage command vector (vd *, vq *) from the current controller 3 of the AC rotating machine control means 31 and receives the detected current vector (id, iq) from the three phases / phases of the AC rotating machine control means 31. The dq-axis secondary magnetic flux vector φrL ^ is received from the secondary magnetic flux calculator 10 of the secondary magnetic flux vector calculator 32. The adaptive observer 11 obtains the estimated speed ωr ^ and the estimated primary angular frequency ω1 from the voltage command vector (vd *, vq *), the detected current vector (id, iq), and the dq-axis secondary magnetic flux vector φrL ^. . The adaptive observer 11 supplies the estimated speed ωr ^ to the speed controller 2 of the AC rotating machine control means 31 and supplies the estimated primary angular frequency ω1 to the integrator 7.

The integrator 7 receives the estimated primary angular frequency ω <b> 1 from the adaptive observer 11. The integrator 7 integrates the estimated primary angular frequency ω1 to obtain the estimated rotational position θ1. The integrator 7 sends the estimated rotational position θ1 to the 3-phase / 2-phase converter 8, the 2-phase / 3-phase converter 4 of the AC rotating machine control means 31, and the subtractor 22 of the secondary magnetic flux vector calculation means 32, respectively. Supply.

  Next, the internal configuration of the adaptive observer 11 will be described with reference to FIG. FIG. 2 is a diagram showing an internal configuration of the adaptive observer 11.

  As shown in FIG. 2, the adaptive observer 11 includes a magnetic flux / current estimator 12, a speed estimator 13, a Kw calculator 14, and an integral addition calculator 15.

  The magnetic flux / current estimator 12 receives the voltage command vector (vd *, vq *) from the current controller 3 of the AC rotating machine control means 31, and receives the detected current vector (id, iq) of 3 of the AC rotating machine control means 31. The dq-axis secondary magnetic flux vector φrL ^ is received from the secondary magnetic flux calculator 10 of the secondary magnetic flux vector calculation means 32. Further, the magnetic flux / current estimator 12 receives the estimated speed ωr ^ from the speed estimator 13.

  The magnetic flux / current estimator 12 calculates the estimated primary magnetic flux vector φs from the voltage command vector (vd *, vq *), the detected current vector (id, iq), the dq-axis secondary magnetic flux vector φrL ^, and the estimated speed ωr ^. ^, The estimated secondary magnetic flux vector φr ^ and the estimated primary current vector is ^ are calculated. Then, the magnetic flux / current estimator 12 calculates a current deviation es between the detected primary current vector is = (id, iq) and the estimated primary current vector is ^. The magnetic flux / current estimator 12 supplies the current deviation es to the speed estimator 13 and also supplies the estimated secondary magnetic flux vector φr ^ to the speed estimator 13 and the integral addition amount calculator 15.

  The speed estimator (proportional integral controller) 13 receives the current deviation es and the estimated secondary magnetic flux vector φr ^ from the magnetic flux / current estimator 12, and receives the dq-axis secondary magnetic flux vector φrL ^ of the secondary magnetic flux vector calculation means 32. Received from the secondary magnetic flux calculator 10. Speed estimator 13 receives addition amount α from integral addition amount calculator 15 and switching gain Kw from Kw calculator 14.

  The speed estimator 13 performs proportional integral control from the current deviation es, the estimated secondary magnetic flux vector φr ^, the dq-axis secondary magnetic flux vector φrL ^, and the switching gain Kw, and adds an amount α to the integral term in the result of the proportional integral control. Are added to obtain the estimated speed ωr ^. The speed estimator 13 supplies the estimated speed ωr ^ to the magnetic flux / current estimator 12 and the Kw calculator 14 and outputs it to the speed controller 2 of the AC rotating machine control means 31.

  The Kw calculator 14 receives the estimated speed ωr ^ from the speed estimator 13. The Kw calculator 14 creates a switching gain Kw based on the estimated speed ωr ^. For example, the Kw calculator 14 switches the switching gain Kw in a binary manner based on the estimated speed ωr ^.

  For example, when the estimated speed ωr ^ increases beyond the threshold ωk2, the Kw calculator 14 switches the switching gain Kw from the first value to the second value. The first value is a value larger than 0, for example, 1. The second value is closer to 0 than the first value, and is 0, for example. For example, the Kw calculator 14 switches the switching gain Kw from the second value to the first value when the estimated speed ωr ^ becomes smaller than the threshold value ωk1.

  Note that the upward threshold ωk2 and the downward threshold ωk1 may be different from each other, for example. For example, as shown in FIG. 4, the threshold value ωk2 may be a value larger than the threshold value ωk1.

  The Kw calculator 14 supplies the switching gain Kw to the integral addition amount calculator 15 and the speed estimator 13.

  The integral addition amount calculator 15 receives the estimated secondary magnetic flux vector φr ^ from the magnetic flux / current estimator 12, receives the switching gain Kw from the Kw calculator 14, and calculates the dq-axis secondary magnetic flux vector φrL ^ as a secondary magnetic flux vector. Received from the secondary magnetic flux calculator 10 of the means 32. The integral addition amount calculator 15 calculates an addition amount α to be added to the integral term of the speed estimator (proportional integral controller) 13 according to the change of the switching gain Kw. The integral addition amount calculator 15 supplies the addition amount α to the speed estimator 13.

  The operation of the present invention will be described below. However, id command calculator 1, speed controller 2, current controller 3, 2-phase / 3-phase converter 4, voltage application means 5, current detection means 6, AC rotating machine M, 3-phase / 2-phase converter 8 Since the rotational position detector 9 is not the essence of the present invention, detailed description of the operation is omitted.

  As shown in FIG. 3, the secondary magnetic flux calculator 10 uses a position deviation Δθ that is a deviation between the detected rotational position θr acquired by the rotational position detector 9 and the estimated rotational position θ1 estimated by the adaptive observer 11. Then, calculations shown in the following formulas 2 and 3 are performed to create a dq-axis secondary magnetic flux vector φrL ^ shown in the following formula 1.

  For example, when an α-axis-β-axis coordinate system is considered as an arbitrary fixed two-axis orthogonal coordinate system, a d-axis secondary magnetic flux φdrL ^ shown in Formula 2, a q-axis secondary magnetic flux φqrL ^ shown in Formula 3, and a formula 1 Each of the dq-axis secondary magnetic flux vectors φrL ^ is expressed as shown in FIG.

  In Equations 2 and 3, φf is an induced voltage constant set in the control device 100 in advance. Further, if there is an estimated rotational position θ1 by the adaptive observation means 33 and a detected rotational position θr by the rotational position detection means 34, the position deviation Δθ and the secondary magnetic flux vector can be obtained, so that means for obtaining θr is the essence of the present invention. is not. As another means for obtaining the detected rotational position θr, for example, there is a position estimation means that uses the saliency of the AC rotating machine M when the AC rotating machine M is a permanent magnet synchronous rotating machine. That is, the rotational position detection means 34 may have a position estimation means that uses the saliency of the AC rotating machine M instead of the rotational position detector 9. Further, when the saliency is used, it is possible to directly obtain the position deviation Δθ. However, the present invention is also applicable to such a case.

  The magnetic flux / current estimator 12 calculates the voltage command vector (vd *, vq *), the detected current vector (id, iq), the dq axis secondary magnetic flux vector φrL ^ by, for example, Formula 4, Formula 5, Formula 8, and Formula 9. And an estimated primary magnetic flux vector φs ^, an estimated secondary magnetic flux vector φr ^, and an estimated primary current vector is ^. Then, the magnetic flux / current estimator 12 calculates the current deviation es between the detected primary current vector is = (id, iq) and the estimated primary current vector is ^ using, for example, Equation 6. At this time, the magnetic flux / current estimator 12 may further calculate the magnetic flux deviation er between the estimated secondary magnetic flux vector φr ^ and the dq-axis secondary magnetic flux vector φrL ^ by using Equation 7, for example.

  Where φs ^ is the estimated primary magnetic flux vector, φr ^ is the estimated secondary magnetic flux vector, vs is the primary voltage vector, is ^ is the estimated primary current vector, is is the detected primary current vector, and R is the primary resistance. , L is the primary winding inductance, and h1, h2, h3, and h4 are adaptive observer gains.

The speed estimator 13 performs the PI control (proportional integral control) shown in Equation 10, and adds the addition amount α shown in Equation 11 to the integral term (term related to 1 / s in Equation 10) in the result of the proportional integral control. The estimated speed ωr ^ is obtained by addition.

  Here, Kap is a PI controller proportional gain, ωapi is a PI controller breakpoint angular frequency, s is a Laplace operator, Kw is a switching gain, and ωr ^ ′ is an estimated speed before the addition amount α is added. It is.

  The Kw calculator 14 creates a switching gain Kw based on the estimated speed ωr ^. FIG. 4 shows the relationship between the switching gain Kw and the estimated speed ωr ^. The threshold value ωk1 is a speed at which switching is performed in the downward direction when the estimated speed ωr ^ decreases. The threshold value ωk2 is a speed at which switching is performed in the upward direction when the estimated speed ωr ^ increases. As shown in FIG. 4, by providing a hysteresis characteristic between the threshold value ωk1 and the threshold value ωk2, the switching gain Kw can be prevented from repeatedly changing at high speed due to the vibration component of the estimated speed ωr ^. The switching operation of Kw can be stabilized.

  The integral addition amount calculator 15 refers to the value of the switching gain Kw and calculates the addition amount α to be added to the PI controller integral term of the acceleration estimator. Formula 11 shows a calculation formula of the compensation amount Δωr to be used for the addition amount α. The compensation amount Δωr corresponds to the amount that changes when the switching gain Kw is switched in Equation 11. That is, the integral addition amount calculator 15 calculates the compensation amount Δωr so as to compensate for the fluctuation of the estimated speed ωr ^ ′ that occurs when the switching gain Kw is switched, and generates the addition amount α using the compensation amount Δωr. And supplied to the speed estimator 13. As a result, the speed estimator 13 performs proportional integral control, and adds the addition amount α to the integral term in the result of the proportional integral control to obtain the estimated speed ωr ^.

  For example, the speed estimator 13 includes a proportional-plus-integral controller 13a, an adder 13b, and a setter 13c. The proportional-plus-integral controller 13a performs proportional-integral control as shown in Equation 10, for example, calculates the estimated speed ωr ^ ', and supplies it to the adder 13b. The setter 13c supplies the set addition amount α to the adder 13b while setting the addition amount α received from the integral addition amount calculator 15 with the responsiveness set in the proportional integration controller 13a, for example. The adder 13b adds the addition amount α to the estimated speed ωr ^ 'to obtain the estimated speed ωr ^.

  For example, when the integral addition amount calculator 15 recognizes that the switching gain Kw is switched from a first value (for example, 1) to a second value (for example, 0), the compensation amount Δωr becomes the acceleration estimator 13. The amount of addition is α = + Δωr so as to be added to the integral term of the PI controller (see FIG. 5). For example, when the integral addition amount calculator 15 recognizes that the switching gain Kw is switched from the second value (for example, 0) to the first value (for example, 1), the compensation amount Δωr is accelerated. The addition amount α = −Δωr so that it is subtracted from the integral term of the PI controller of the estimator 13 (see FIG. 5).

  FIG. 5 shows the operation of the main variable at the time of acceleration / deceleration according to the present embodiment with the horizontal axis as time. For each section, numbers TP1 to TP5 are given for the explanation of the operation. TP1 is an interval from acceleration from the stop until the switching gain Kw is switched from a first value (for example, 1) to a second value (for example, 0). TP2 is a section from when the switching gain Kw is switched to the second value (for example, 0) until the acceleration is finished. TP3 is a section from the end of acceleration to the start of deceleration. TP4 is a section from when deceleration starts until Kw switches from a second value (for example, 0) to a first value (for example, 1). TP5 is a section from when the switching gain Kw is switched to a first value (for example, 1) to when it stops.

  For example, the operation during acceleration when the switching gain Kw is switched from a first value (for example, 1) to a second value (for example, 0) is as follows. When shifting from the section TP1 to the section TP2, the addition amount α = + Δωr corresponding to the compensation amount Δωr expressed by the equation 11 is added to the integral term. Thereby, the continuity of the estimated speed ωr ^ can be maintained as shown in FIG. The integral term is settled with the responsiveness set in the PI controller for speed estimation after the interval TP2. For example, in the acceleration estimator 13, the integral term is a value that gradually approaches 0 from + Δωr (for example, with the responsiveness set in the PI controller) as the addition amount α corresponding to the compensation amount Δωr shown in Equation 11. Is added.

  For example, the operation at the time of deceleration when the switching gain Kw is switched from the second value (for example, 0) to the first value (for example, 1) is as follows. When shifting from the section TP4 to the section TP5, the addition amount α = −Δωr corresponding to the compensation amount Δωr expressed by the equation 11 is added to the integral term. Thereby, the continuity of the estimated speed ωr ^ can be maintained as shown in FIG. The integral term is settled with the responsiveness set in the PI controller for speed estimation after the interval TP5. For example, in the acceleration estimator 13, the integral term gradually approaches 0 from −Δωr (for example, with the responsiveness set in the PI controller) as the addition amount α corresponding to the compensation amount Δωr shown in Equation 11. Add some value.

  With such an action, discontinuity in speed estimation when switching gain Kw is switched can be suppressed, and smooth acceleration / deceleration can be realized.

Here, suppose that the adaptive observation means 33 does not have the integral addition amount calculator 15. In this case, for example, when the speed estimator 13 calculates the estimated speed ωr ^ ′ in the adaptive observation means 33, the switching gain Kw is switched from the first value (for example, 1) to the second value (for example, 0). At the time of acceleration, for example, as shown in Equation 10, the term including the dq-axis secondary magnetic flux vector φrL ^ disappears instantaneously, and the estimated speed ωr ^ 'from the interval TP1 to the interval TP2 as indicated by a one-dot chain line in FIG. Discontinuities can increase. Alternatively, for example, when the speed estimator 13 calculates the estimated speed ωr ^ ′ in the adaptive observation means 33, the acceleration at which the switching gain Kw is switched from the second value (for example, 0) to the first value (for example, 1). Occasionally, a term including the dq-axis secondary magnetic flux vector φrL ^ that has disappeared appears instantaneously as shown in Equation 10, for example, and the estimated speed ωr from the interval TP4 to the interval TP5 as indicated by a dashed line in FIG. The discontinuity of ^ 'may increase.

  On the other hand, in the first embodiment, in the control device 100 for the AC rotating machine M, the adaptive observation means 33 includes the integral addition amount calculator 15. The integral addition amount calculator 15 and the speed estimator 13 compensate for fluctuations in the estimated speed that occur when the switching gain is switched. For example, the integral addition amount calculator 15 calculates the compensation amount Δωr so as to compensate for the fluctuation of the estimated speed ωr ^ ′ that occurs when the switching gain Kw is switched, and generates the addition amount α using the compensation amount Δωr. And supplied to the speed estimator 13. As a result, the speed estimator 13 can perform proportional integral control and add the addition amount α to the integral term in the result of the proportional integral control to obtain the estimated speed ωr ^. Can be suppressed. Therefore, the speed estimation discontinuity at the time of switching the switching gain Kw can be eliminated, and smooth acceleration / deceleration can be realized.

  In the first embodiment, in the control device 100 of the AC rotating machine M, the adaptive observation means 33 estimates the speed so as to eliminate the detected current vector is = (id, iq) and the deviation vector es of the estimated current vector is ^. The speed is estimated by proportional-integral control by a controller (proportional-integral controller) 13. At this time, the integral addition amount calculator 15 calculates the compensation amount Δωr so as to compensate for the fluctuation of the estimated speed ωr ^ ′ that occurs when the switching gain Kw is switched, and uses the compensation amount Δωr to calculate the addition amount α. Generated and supplied to the speed estimator 13. The speed estimator 13 adds the fluctuation amount of the estimated speed to the integral term of the proportional-plus-integral controller 13a, and settles with the responsiveness set in the proportional-plus-integral controller 13a. Thereby, the continuity of the estimated speed ωr ^ at the time of switching the switching gain Kw can be improved.

Embodiment 2. FIG.
Next, the control device 200 for the AC rotating machine M according to the second embodiment will be described. Below, it demonstrates focusing on a different part from Embodiment 1. FIG.

  In the first embodiment, the addition amount α calculated by the integral addition amount calculator 15 is settled with the responsiveness set in the proportional-plus-integral controller 13a. However, in the second embodiment, the addition amount α itself is arbitrarily set. Provide attenuation characteristics.

  Specifically, the adaptive observation means 233 of the control device 200 includes an adaptive observer 211 as shown in FIG. 6 instead of the adaptive observer 11 (see FIG. 2). The adaptive observer 211 includes a speed estimator 213 instead of the speed estimator 13 (see FIG. 2), and further includes an adder 218 and an attenuator 219. FIG. 6 is a diagram showing an internal configuration of the adaptive observer 211.

  For example, the speed estimator 213 does not include the adder 13b and the setter 13c. For example, the proportional-integral controller 13 a of the speed estimator 213 performs proportional-integral control, and supplies the obtained estimated speed ωr ^ ′ to the adder 218.

  The attenuator 219 gives an arbitrary attenuation characteristic to the addition amount α obtained by the integral addition amount calculator 15, and outputs the attenuated addition amount α ′ to the adder 218.

  The adder 218 adds the estimated speed ωr ^ ′ and the attenuated addition amount α ′ to create the estimated speed ωr ^. The adder 218 supplies the estimated speed ωr ^ to the magnetic flux / current estimator 12 and the Kw calculator 14 and outputs it to the speed controller 2 of the AC rotating machine control means 31.

  For example, the operation when the attenuator 219 is a first-order lag element represented by Expression 12 will be described. FIG. 7 shows the operation of the addition amount α ′ when the horizontal axis is time. The addition amount α is given to the attenuator 219 at the switching time Tk, and thereafter the attenuator 219 attenuates the addition amount α with the characteristic shown in Equation 12. The addition amount α ′ attenuated by the attenuator 219 is added to the output of the speed estimator 213 (that is, the estimated speed ωr ^ ′) to create the estimated speed ωr ^.

  In Equation 12, τ is a time constant and s is a Laplace operator.

  Alternatively, for example, an operation when the attenuator 219 has a linear attenuation characteristic will be described. FIG. 8 shows the operation of the attenuated addition amount α ′ when the attenuator 219 has a linear attenuation characteristic. The addition amount α is given to the attenuator 219 at the switching time Tk, and then the attenuator 219 linearly attenuates the addition amount α to 0 with an arbitrary attenuation time β. The addition amount α ′ attenuated by the attenuator 219 is added to the output of the speed estimator 213 (that is, the estimated speed ωr ^ ′) to create the estimated speed ωr ^.

  By such an action, discontinuity of speed estimation at the time of switching of the switching gain Kw can be eliminated, and smooth acceleration / deceleration can be realized.

  As described above, in the second embodiment, in the control device 200 for the AC rotating machine M, the adaptive observation unit 233 includes the attenuator 219 that attenuates the input signal with an arbitrary characteristic. The attenuator 219 attenuates the addition amount α with an arbitrary attenuation characteristic and supplies it to the adder 218. The adder 218 adds the addition amount α ′ attenuated by the attenuator 219 to the estimated speed ωr ^ ′ calculated by the speed estimator 213 to obtain the estimated speed ωr ^. Thereby, the discontinuity of the estimated speed ωr ^ can be further suppressed.

  In the second embodiment, in the adaptive observation means 233, the attenuator 219 has, for example, a first-order lag characteristic. Thus, the estimated speed ωr ^ can be obtained by adding the addition amount α ′ attenuated by the first-order lag characteristic to the estimated speed ωr ^ ′ calculated by the speed estimator 213. Therefore, when the discontinuity of the estimated speed ωr ^ corresponds to the first-order lag characteristic, the discontinuity of the estimated speed ωr ^ can be effectively suppressed.

  In the second embodiment, in the adaptive observation unit 233, the attenuator 219 has, for example, a linear attenuation characteristic. Thus, the estimated speed ωr ^ can be obtained by adding the amount of addition α ′ attenuated by the linear attenuation characteristic by the attenuator 219 to the estimated speed ωr ^ ′ calculated by the speed estimator 213. Therefore, the attenuator 219 can be configured easily.

Embodiment 3 FIG.
Next, the control device 300 for the AC rotating machine M according to the third embodiment will be described. Below, it demonstrates focusing on a different part from Embodiment 1. FIG.

  In the first embodiment, the addition amount α calculated by the integral addition amount calculator 15 is added to the integral term of the proportional integral controller 13a of the speed estimator 13 to ensure the continuity of the estimated speed ωr ^. However, in the third embodiment, continuity is imparted to the switching gain Kw itself.

  Specifically, the adaptive observation means 333 of the control device 300 includes an adaptive observer 311 as shown in FIG. 9 instead of the adaptive observer 11 (see FIG. 2). The adaptive observer 311 includes a speed estimator 313 instead of the speed estimator 13 (see FIG. 2), does not include the integral addition amount calculator 15, and further includes a continuous characteristic adder 320. FIG. 9 is a diagram illustrating an internal configuration of the adaptive observer 311.

  The continuous characteristic imparting unit 320 imparts a continuous characteristic for continuously changing the input switching gain Kw in terms of time, and its output is a switching gain Kwfil to which the continuous characteristic is imparted. The continuous characteristic imparted to the switching gain Kw by the continuous characteristic imparting device 320 may be any characteristic that provides temporal continuity to the switching gain Kw. Examples include the first-order lag characteristic shown in Formula 12 and FIG. 7, and the characteristic that changes linearly in proportion to time from the switching time of the switching gain Kwfil as shown in FIG.

  Further, the speed estimator 313 does not include, for example, the adder 13b and the setter 13c. For example, the proportional-plus-integral controller 313a of the speed estimator 313 performs proportional-integral control shown in, for example, Expression 13 to obtain the estimated speed ωr ^. Formula 13 is different from Formula 10 in that the switching gain Kw is changed to the switching gain Kwfil.

  FIG. 10 shows the operation of main variables when the characteristic of the continuous characteristic assigner 320 is the first-order lag characteristic. At the switching point where the value of the switching gain Kw changes stepwise, the switching gain Kwfil to which the continuous characteristic is given continuously changes with the first order lag characteristic. Thereby, the term including the dq-axis secondary magnetic flux vector φrL ^ in Expression 13 can be continuously changed.

  As a result, the estimated speed corresponding to the dq-axis secondary magnetic flux vector φrL ^ can be continuously changed, so that the discontinuity of speed estimation at the time of switching of the switching gain Kw can be eliminated, and smooth acceleration / deceleration can be achieved. Can be realized.

  As described above, in the third embodiment, in the control device 300 of the AC rotating machine M, the adaptive observation unit 333 includes the continuous characteristic imparting unit 320 that imparts temporal continuous characteristics to the input signal. The continuous characteristic imparting unit 320 imparts a temporal continuous characteristic to the switching gain Kw, and supplies the switching gain Kwfil to which the continuous characteristic is imparted to the speed estimator 313. Thereby, the proportional-plus-integral controller 313a of the speed estimator 313 can perform the proportional-integral control using the switching gain Kwfil to which the continuous characteristic is given, and can obtain the estimated speed ωr ^. As a result, the discontinuity of the estimated speed ωr ^ can be suppressed.

  In the third embodiment, in the control device 300 for the AC rotating machine M, the characteristic of the continuous characteristic assigner 320 of the adaptive observation unit 333 is, for example, a first-order lag characteristic. As a result, proportional integral control is performed using the switching gain Kwfil to which the continuous characteristic of the first-order lag characteristic is given, and the estimated speed ωr ^ can be obtained. Therefore, when the discontinuity of the estimated speed ωr ^ corresponds to the first-order lag characteristic, the discontinuity of the estimated speed ωr ^ can be effectively suppressed.

  In the third embodiment, in the control device 300 for the AC rotating machine M, the characteristic of the continuous characteristic assigner 320 of the adaptive observation unit 333 is, for example, a linear characteristic. Accordingly, the proportional integral control is performed using the switching gain Kwfil to which the continuous characteristic of the linear characteristic is given by the continuous characteristic giving unit 320, and the estimated speed ωr ^ can be obtained. Therefore, the continuous characteristic imparting device 320 can be configured easily.

  As described above, the control device for an AC rotating machine according to the present invention is useful for controlling the AC rotating machine.

  5 Voltage application means, 31 AC rotating machine control means, 32 Secondary magnetic flux vector calculation means, 33, 233, 333 Adaptive observation means, 100, 200, 300 Controller, M AC rotating machine.

Claims (7)

AC rotating machine control means for creating a voltage command vector so that the detected current vector matches the current command vector using the AC rotating machine position information;
Voltage applying means for applying a voltage to the AC rotating machine based on the voltage command vector;
Secondary magnetic flux vector computing means for computing a secondary magnetic flux vector in the AC rotating machine;
Adaptive observation means for obtaining and outputting an estimated rotational position and an estimated speed according to the voltage command vector, the detected current vector, and the secondary magnetic flux vector;
With
The adaptive observation means includes a magnetic flux / current estimator having an adaptive observer gain, a speed estimator, a switching gain calculator, and a compensation means having a continuous property assigner,
The magnetic flux / current estimator calculates an estimated primary magnetic flux vector, an estimated secondary magnetic flux vector, and an estimated primary current vector based on the voltage command vector, the detected current vector, and the secondary magnetic flux vector,
The adaptive observer gain amplifies a current deviation vector which is a deviation between the estimated primary current vector and the detected current vector, and a magnetic flux deviation vector which is a deviation between the estimated secondary magnetic flux vector and the secondary magnetic flux vector. And
The speed estimator calculates an estimated speed from the current deviation vector, the estimated secondary magnetic flux vector, and the secondary magnetic flux vector,
The switching gain calculator is generated based on the estimated speed, and calculates the switching gain so that the switching gain is switched with hysteresis characteristics after the acceleration of the AC rotating machine is started.
The adaptive observation means can switch the use or non-use of the secondary magnetic flux vector with the switching gain,
It said compensation means, a variation in the estimated speed of the switching gain occurs when switched to compensation,
The continuous characteristic applying instrument, the switching gain are input, continuous characteristics Outputs switching gain granted to impart temporal continuity characteristic to the switching gain,
The estimated speed for the secondary magnetic flux vector is given a temporal continuity characteristic by the switching gain to which the continuous characteristic is given, the temporal continuity characteristic is not given to the adaptive observer gain, and the switching gain is A control device for an AC rotating machine, characterized in that a temporal continuous characteristic is imparted by the continuous characteristic imparting device. The control device for an AC rotating machine according to claim 1, wherein the characteristic of the continuous characteristic imparting device is a first-order lag characteristic. The control device for an AC rotating machine according to claim 1, wherein the characteristic of the continuous characteristic imparting device is a linear characteristic. AC rotating machine control means for creating a voltage command vector so that the detected current vector matches the current command vector using the AC rotating machine position information;
Voltage applying means for applying a voltage to the AC rotating machine based on the voltage command vector;
Secondary magnetic flux vector computing means for computing a secondary magnetic flux vector in the AC rotating machine;
Adaptive observation means for obtaining and outputting an estimated rotational position and an estimated speed according to the voltage command vector, the detected current vector, and the secondary magnetic flux vector;
With
The adaptive observation means can switch whether or not the secondary magnetic flux vector is used with a switching gain,
The adaptive observation means includes a compensation means for compensating for an estimated speed fluctuation occurring when the switching gain is switched.
The adaptive observation means includes
Proportional integral control with a proportional integral controller to eliminate the deviation vector of the detected current vector and the estimated current vector to estimate the speed,
The compensation means, when adding an addition amount corresponding to the fluctuation amount of the estimated speed to the integral term of the proportional integration controller, settling the addition amount with the responsiveness set in the proportional integration controller A control device for an AC rotating machine, characterized in that it has means. AC rotating machine control means for creating a voltage command vector so that the detected current vector matches the current command vector using the AC rotating machine position information;
Voltage applying means for applying a voltage to the AC rotating machine based on the voltage command vector;
Secondary magnetic flux vector computing means for computing a secondary magnetic flux vector in the AC rotating machine;
Adaptive observation means for obtaining and outputting an estimated rotational position and an estimated speed according to the voltage command vector, the detected current vector, and the secondary magnetic flux vector;
With
The adaptive observation means can switch whether or not the secondary magnetic flux vector is used with a switching gain,
The adaptive observation means includes a compensation means for compensating for an estimated speed fluctuation occurring when the switching gain is switched.
The adaptive observation means includes an attenuator that attenuates an input signal with an arbitrary characteristic,
The control device for an AC rotating machine, wherein the compensation means includes addition means for adding an addition amount obtained by attenuating an addition amount corresponding to the estimated speed fluctuation amount by the attenuator to the estimation speed. The control device for an AC rotating machine according to claim 5, wherein the attenuator has a first-order lag characteristic. The control device for an AC rotating machine according to claim 5, wherein the attenuator has a linear attenuation characteristic.

Images