JP2014007851A - Control device for electric motor, and control method for electric motor - Google Patents

Abstract

The temperature of a permanent magnet is accurately estimated even when the temperature and magnetic flux density of the permanent magnet do not correspond one-to-one.
A control device and a control method for an electric motor 100 detect fundamental waves id, iq and n-order harmonics id_n, iq_n of the electric motor 100, respectively, and determine a magnetic flux density by a permanent magnet from the fundamental waves id, iq of the electric motor 100. The fundamental wave component Φa_1st is obtained, and the nth harmonic component Φa_nst of the magnetic flux density by the permanent magnet is obtained from the nth harmonics id_n and iq_n of the electric motor 100. With reference to data indicating the correlation between the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density and the temperature Tm of the permanent magnet, the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density The temperature Tm or the magnetic flux density Φa is estimated.
[Selection] Figure 1

Description

  The present invention relates to an electric motor control device including a permanent magnet and an electric motor control method.

  Generally, a permanent magnet has temperature dependence, and the magnetic flux of a permanent magnet decreases as the temperature of the permanent magnet increases. In addition, the permanent magnet is vulnerable to heat, and the demagnetization resistance decreases as the temperature increases. Therefore, when the temperature of the permanent magnet changes, the accuracy of the torque output from the electric motor in which the permanent magnet is embedded, the control efficiency of the electric motor, and the current response are deteriorated. Moreover, in order to suppress demagnetization of the permanent magnet due to temperature rise, it is necessary to make a sufficiently large heat-resistant margin and to manufacture the permanent magnet using a magnet material having high heat resistance, which increases the manufacturing cost. If the temperature of the permanent magnet can be known, the torque accuracy of the motor, the control efficiency of the motor, and the current response can be improved, and the design margin of the permanent magnet can be reduced. Thus, an invention for estimating the temperature of a permanent magnet embedded in a rotor of an electric motor has been proposed (for example, see Patent Document 1).

JP 2008-43128 A

  In patent document 1, the temperature of a permanent magnet is estimated from the induced voltage in a steady state. The electric motor described in Patent Document 1 has a double rotor structure including an outer rotor and an inner rotor, and has a variable mechanism that changes the magnetic flux density of the entire rotor in accordance with the phase difference between the rotors. Therefore, in an electric motor having such a variable magnetic flux density mechanism, if the phase difference between the rotors is not measured, whether the induced voltage (magnetic flux density) changes due to the variable mechanism or the temperature change of the permanent magnet. Cannot judge whether it is due to.

  The present invention has been made in view of the above problems, and the purpose thereof is to estimate the temperature of the permanent magnet even when the temperature and magnetic flux density of the permanent magnet do not correspond one-to-one. It is providing the control apparatus of an electric motor, and the control method of an electric motor.

  In order to achieve the above object, an electric motor control apparatus and control method according to an embodiment of the present invention detects a fundamental wave and an nth harmonic of the electric motor, respectively, and generates a fundamental wave of magnetic flux density by a permanent magnet from the fundamental wave of the electric motor. The component is obtained, and the nth harmonic component of the magnetic flux density by the permanent magnet is obtained from the nth harmonic of the electric motor. Then, referring to the data indicating the correlation between the fundamental wave component and the nth harmonic component of the magnetic flux density and the temperature of the permanent magnet, the temperature or magnetic flux of the permanent magnet is obtained from the fundamental wave component and the nth harmonic component of the magnetic flux density. Estimate density.

  According to the motor control device and motor control method of the present invention, the temperature of the permanent magnet can be accurately estimated even when the temperature of the permanent magnet and the magnetic flux density do not correspond one-to-one.

FIG. 1A is a block diagram showing a part of a control device for controlling the electric motor of FIG. 2, and FIG. 1B is a remaining one of the control device for the electric motor according to the first embodiment of the present invention. It is a block diagram which shows a part. FIG. 2 is a partial cross-sectional view showing a first example of the structure of one pole of the permanent magnet type motor according to the first embodiment of the present invention. FIG. 3 is a graph showing an example of data indicating the correlation between the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density and the temperature Tm of the permanent magnets 10a and 10b in the control device of FIG. FIG. 4 is a partial cross-sectional view showing a second example of the structure of the electric motor in the first embodiment of the present invention. FIG. 5 is a graph showing an example of data indicating the correlation between the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density and the temperature Tm of the permanent magnets 10a and 10b in the electric motor of FIG. FIG. 6A is a partial cross-sectional view showing a third example of the structure of the electric motor according to the first embodiment of the present invention, and FIG. 6B shows the rotating shaft 9 of the electric motor of FIG. FIG. FIG. 7A is a graph showing an example of data indicating the correlation between the phase fundamental wave component and the nth-order harmonic component and the phase difference (skew) in the electric motors of FIGS. 6A and 6B. It is. FIG. 7B is a graph showing an example of data indicating a correlation between the fundamental wave component Φa_1st of the magnetic flux density and the temperature Tm of the permanent magnets 10a and 10b in the electric motors of FIGS. 6A and 6B. is there. FIG. 8 is a flowchart showing an example of a procedure for estimating the magnetic flux density and the temperature of the permanent magnets 10a and 10b for the electric motor 100 having a variable mechanism that changes the magnetic flux density of the permanent magnets 10a and 10b as the motor constant. FIG. 9 is a graph for explaining a method of calculating the magnetic flux density distribution in the circumferential direction of the electric motor 100 from the magnetic flux density (amplitude), the fundamental wave component of the phase, and the nth harmonic component. FIG. 10 is a flowchart illustrating an example of a procedure for estimating the induced voltage and the temperature Tm of the permanent magnets 10a and 10b for the motor 100 having a variable mechanism that changes the induced voltage generated in the armature coil 6 as the motor constant. FIG. 11 is a flowchart illustrating an example of a procedure for estimating the inductance and the temperature Tm of the permanent magnets 10a and 10b for the motor 100 having a variable mechanism that changes the inductance of the armature coil 6 as the motor constant. FIG. 12 is a flowchart showing a temperature estimation method when detecting fundamental waves id and iq of current flowing in the armature coil 6 and n-order harmonics id_n and iq_n as fundamental waves and n-order harmonics of the electric motor 100. . FIG. 13 is a flowchart showing a temperature estimation method when detecting fundamental waves vd and vq of voltage applied to the armature coil 6 and n-order harmonics vd_n and vq_n as fundamental waves and n-order harmonics of the electric motor 100. It is. FIG. 14 is a flowchart showing a temperature estimation method when detecting the fundamental wave and the nth harmonic of the inductance of the armature coil 6 as the fundamental wave and the nth harmonic of the electric motor 100. FIG. 15 is a flowchart illustrating a temperature estimation method in the case of detecting the fundamental wave and the nth harmonic of the torque generated in the motor 100 as the fundamental wave and the nth harmonic of the motor 100. FIG. 16 is a flowchart showing a temperature estimation method in the case of detecting the fundamental wave and the nth harmonic of the position of the rotor 4 as the fundamental wave and the nth harmonic of the electric motor 100. FIG. 17 is a flowchart showing a temperature estimation method in the case of detecting the fundamental wave and the nth harmonic of the current flowing through the armature coil 6 and the voltage applied to the armature coil 6. FIG. 18 shows an example of a procedure for estimating the motor constant and the magnet temperature by combining the position estimation error of the rotor 4 and the detected value of the current flowing through the armature coil 6 and the voltage applied to the armature coil 6. It is a flowchart to show. FIG. 19 is a flowchart illustrating an example of a procedure for estimating the temperature Tm of the permanent magnets 10a and 10b in the electric motor 100 that does not include the magnetic flux density variable mechanism. FIG. 20 is a flowchart illustrating an example of a temperature estimation method according to the twelfth embodiment. FIG. 21 is a flowchart showing another example of the temperature estimation method according to the twelfth embodiment.

  Embodiments of the present invention will be described below with reference to the drawings. In the description of the drawings, the same parts are denoted by the same reference numerals.

(First embodiment)
FIG. 2 is a partial cross-sectional view showing a first example of the structure of one pole of a permanent magnet type electric motor (hereinafter referred to as “electric motor”) in an embodiment of the present invention. The electric motor includes a stator 2 fixed to the outer peripheral case, and a rotor 4 disposed on the inner peripheral side of the stator 2 via an air gap and fixed to the rotor shaft 3. The stator 2 has a stator core formed by laminating electromagnetic steel plates, for example, and has a structure in which armature coils 6 are wound in slots 5 provided at equal intervals in the circumferential direction of the stator core.

  On the other hand, the rotor 4 has a rotor core 7 formed by laminating electromagnetic steel plates around the axis of the rotor shaft 3, for example, and permanent magnets 10a and 10b that form magnetic poles at equal intervals in the circumferential direction of the rotor core 7 are arranged. Has a structured. Permanent magnets 10a and 10b have a configuration in which permanent magnets 10a with N poles facing stator 2 and permanent magnets 10b with S poles are alternately arranged in the circumferential direction. In the electric motor configured as described above, the rotating magnetic field generated by energizing the armature coil 6 of the stator 2 with alternating current having a driving frequency corresponding to the number of pole pairs of the rotor 4, and the permanent magnets 10 a and 10 b of the rotor 4. The rotor 4 and the rotor shaft 3 are rotated by the interaction with the magnet magnetic field generated by.

  In the electric motor shown in FIG. 2, each of the permanent magnets 10a and 10b forming the magnetic poles of the rotor 4 has a relatively large coercive force (hereinafter referred to as high coercive force magnet portions 11a and 11b) and a relatively coercive force. It has a configuration in which small magnets (hereinafter referred to as low coercive force magnet portions 12a and 12b) are combined. A composite current obtained by superimposing the harmonic component of the drive frequency on the fundamental component of the drive frequency under the control of the motor control device described later with reference to FIGS. 1 (a) and 1 (b). Is applied to the armature coil 6, and the demagnetization state of the low coercivity magnet portions 12a and 12b of the permanent magnets 10a and 10b is controlled by the harmonic component of the composite current, thereby controlling the magnetic flux density by the permanent magnets 10a and 10b. To do. As described above, the electric motor shown in FIG. 2 has a variable magnetic flux density mechanism by controlling the demagnetization state of the low coercive force magnet portions 12a and 12b. As the high coercive force magnet portions 11a and 11b, for example, NdFeB magnets can be used. Further, as the low coercive force magnet portions 12a and 12b, a magnet having a high magnetomotive force and a relatively small coercive force, for example, an alnico magnet or a neodymium magnet not added with an element such as Dy for increasing the coercive force is used. Can do.

  First, the configuration of the motor control device according to the first embodiment of the present invention will be described with reference to FIGS. 1 (a) and 1 (b). The electric motor 100 is connected to a DC power source 102 via an inverter 101. The control device is configured to drive the electric motor 100 including the rotor 4 having the permanent magnets 10a and 10b and the stator 2 having the armature coil 6 to the fundamental frequency of driving frequency and n (n is a natural number of 2 or more) order harmonics. To control.

  The control device applies voltage command values for U, V, and W phases to the inverter 101 to control the switching operation of each phase arm of the inverter 101, thereby providing poles to the armature coils 6 of each phase of the electric motor 100. An alternating current having a driving frequency corresponding to the logarithm is energized to control the operation of the electric motor 100. The current sensor 103 detects currents iu, iv, and iw that are supplied to the armature coil 6. The position sensor 104 detects the rotation angle θ as the position of the rotor 4. The position sensor 104 differentiates the rotation angle θ of the rotor 4 to calculate the electrical angular velocity ω of the rotor 4. The value detected by the current sensor 103 is fed back to a first control block 30 and a second control block 40 described later. Further, the calculated value by the position sensor 104 is fed back to the first control block 30.

  The control device includes harmonic components for controlling the magnetization state of the first control block 30 for current control corresponding to the fundamental wave of the driving frequency of the electric motor 100 and the low coercivity magnet portions 12a and 12b included in the rotor 4. Are provided with a second control block 40 for current control and an adder 50.

In the first control block 30, the current command id ^* for fundamental wave control corresponding to the required torque for the electric motor 100 and the difference between the current command iq ^* and the feedback signal are input to the current controller 31. The current controller 31 generates a voltage command to be applied to the armature coil 6. The voltage command generated by the current controller 31 is added to the voltage command generated by the non-interference control unit 34 to become voltage commands vd and vq. Then, the current vector control unit 32a performs coordinate conversion of the voltage commands vd and vq into voltage commands vu, vv, and vw for U, V, and W phases, and outputs them as voltage commands corresponding to the fundamental wave. The current vector control unit 32b converts the fundamental waves of the detection values iu, iv, and iw of the current sensor 103 into the dq coordinate system. The low pass filter 33 removes a high frequency component (noise component) from the detection values id and iq of the fundamental wave converted into the dq coordinate system, and generates the above-described feedback signal.

On the other hand, the second control block 40, the operating state of the motor 100 current control a difference between the current command ID_n ^* and current command Iq_n ^* and the feedback signal for the n-order harmonic component control according to (rpm) 41 To enter. The current controller 41 generates voltage commands vd_n and vq_n to be applied to the armature coil 6. The current vector control unit 42a performs coordinate conversion of the voltage commands vd_n, vq_n generated by the current controller 41 into voltage commands vu_n, vv_n, vw_n for the U, V, and W phases, and the nth-order harmonics. Output as a voltage command corresponding to the component. The current vector control unit 42b converts the nth-order harmonics of the detection values iu, iv, iw of the current sensor 103 into detection values id_n, iq_n of the dq coordinate system. The low-pass filter 43 removes the fundamental wave component and other order frequency components (noise components) from the detected values id_n and iq_n of the nth-order harmonics converted into the dq coordinate system, and generates the aforementioned feedback signal.

  Here, “n” is a natural number of 2 or more, and is changed according to the characteristics of the electric motor 100 as described later. In the present embodiment, only one control block (second control block 40) is shown as a control block for the n-th order high-frequency component of the drive frequency, but the control device has two or more high-frequency components having different orders (n). Component control blocks (third control block, fourth control block,...) May be provided.

  In FIG. 1A, the current vector control unit 32b that converts the fundamental waves of the detection values iu, iv, and iw of the current sensor 103 into the dq coordinate system is an example of “a fundamental wave detection unit that detects a fundamental wave of an electric motor”. It is. Further, the current vector control unit 42b that converts the nth-order harmonics of the detection values iu, iv, and iw of the current sensor 103 into the dq coordinate system is the “nth-order harmonic detection unit that detects the nth-order harmonics of the motor”. It is an example.

  As shown in FIG. 1B, the control device according to the present embodiment includes a fundamental wave magnetic flux computation unit 61 that computes a fundamental wave component Φa_1st of the magnetic flux density by the permanent magnets 10a and 10b from the fundamental wave of the electric motor 100, and the electric motor. An n-order harmonic magnetic flux calculator 62 for calculating an n-order harmonic component Φa_nst of the magnetic flux density from 100 n-order harmonics, a fundamental wave component Φa_1st and an n-order harmonic component Φa_nst of the magnetic flux density, a temperature Tm of the permanent magnet, and A correlation data storage unit 71 that stores data indicating the correlation with the variable amount of the magnetic flux density, and a reference to the data stored in the correlation data storage unit 71, the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density And a magnet parameter estimating unit 70 for estimating the temperature Tm or the magnetic flux density Φa of the permanent magnets 10a and 10b.

  The “fundamental wave of the motor 100” includes the current flowing through the armature coil 6, the voltage applied to the armature coil 6, the inductance of the armature coil 6, the torque generated in the motor 100, and the position of the rotor 4. Includes fundamental waves. The “fundamental wave of the current flowing through the armature coil 6” includes the detected values id and iq of the fundamental wave converted into the dq coordinate system by the current vector control unit 32b. The “basic wave of the voltage applied to the armature coil 6” includes the voltage commands vd and vq generated by the current controller 31. The fundamental wave of the inductance of the armature coil 6, the torque generated in the electric motor 100, and the position of the rotor 4 is derived from the fundamental wave of the current flowing through the armature coil 6 and the fundamental wave of the voltage applied to the armature coil 6. It can be calculated by a known calculation method. Alternatively, the fundamental wave of the position of the rotor 4 can be obtained from the rotation angle θ of the rotor 4 detected by the position sensor 104.

The fundamental wave magnetic flux calculation unit 61 can calculate the fundamental wave component Φa_1st of the magnetic flux density by the permanent magnets 10a and 10b from the fundamental wave of the electric motor 100 by a known calculation method. For example, the magnetic flux density Φa_1st can be calculated by substituting the detected currents id and iq and the voltage commands vd and vq into the equation (1). In equation (1), v _q represents the q-axis voltage, R (t) represents the resistance of the armature coil 6, t represents the temperature of the armature coil 6, i _q represents the q-axis current, and ω represents the angular velocity of the rotor 4, .PHI.a represents a magnetic flux density, L _d represents the inductance of the armature coils 6, i _d represents a current of d axis.

  The “nth harmonic of the motor 100” includes the current flowing through the armature coil 6, the voltage applied to the armature coil 6, the inductance of the armature coil 6, the torque generated in the motor 100, and the rotor 4 The nth harmonic of the position is included. The “nth harmonic of the current flowing through the armature coil 6” includes the detected values id_n and iq_n of the nth harmonic converted into the dq coordinate system by the current vector control unit 42b. The “nth harmonic of the voltage applied to the armature coil 6” includes the voltage commands vd_n and vq_n generated by the current controller 41. The inductance of the armature coil 6, the torque generated in the motor 100, and the nth harmonic of the position of the rotor 4 are the nth harmonic of the current flowing through the armature coil 6 and the voltage applied to the armature coil 6. Can be calculated by a known calculation method from the n-th harmonic. Alternatively, the nth harmonic of the position of the rotor 4 can be obtained from the rotation angle θ of the rotor 4 detected by the position sensor 104.

  The n-order harmonic magnetic flux calculation unit 62 can calculate the n-order harmonic component Φa_nst of the magnetic flux density by the permanent magnets 10a and 10b from the n-order harmonics of the electric motor 100 by a known calculation method. For example, the magnetic flux density Φa_nst can be calculated by substituting the current detection values id_n and iq_n and the voltage commands vd_n and vq_n into the equation (1).

  The correlation data storage unit 71 and the magnet parameter estimation unit 70 will be described later with reference to FIG.

  Referring to FIG. 3, the correlation between the fundamental component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density, the temperature Tm of the permanent magnets 10a and 10b, and the variable amount of the magnetic flux density, stored in the correlation data storage unit 71. An example of data indicating the above will be described. The graph shown in FIG. 3 shows a state in which the low coercive force magnet portions 12a and 12b shown in FIG. 2 are demagnetized to 50% (L) and a state in which the low coercive force magnet portions 12a and 12b are not demagnetized (H ), The magnetic flux densities of the permanent magnets 10a and 10b when the temperatures of the high coercive force magnet portions 11a and 11b are respectively changed to 60 ° C., 100 ° C. and 140 ° C. are expressed for each order by Fourier series expansion. It is a thing. The horizontal axis indicates the order, and the vertical axis indicates the magnetic flux density. The horizontal axis indicates only odd-order components of the primary (fundamental wave), the third order, the fifth order,... That form the magnetic flux density distribution (rectangular wave). In FIG. 3, “L” indicates a state in which the low coercive force magnet portions 12a and 12b are demagnetized to 50%, and “H” indicates a state in which the low coercive force magnet portions 12a and 12b are not demagnetized. .

  As shown in FIG. 3, the fifth harmonic component of the magnetic flux density is less dependent on the temperature of the high coercive force magnet portions 11a and 11b than the fundamental wave and other order harmonic components, and the low coercive force magnet. The demagnetization state of the parts 12a and 12b, that is, the dependence on the variable amount of the magnetic flux density is high. On the other hand, the fundamental wave component of the magnetic flux density is highly dependent on the temperature of the high coercive force magnet portions 11a and 11b as compared with the harmonic component. Therefore, for example, if the nth-order harmonic magnetic flux calculation unit 62 calculates the fifth-order harmonic component Φa_5st of the magnetic flux density, the magnet parameter estimation unit 70 refers to the data shown in FIG. The demagnetization state (variable amount of magnetic flux density) of 12a and 12b can be estimated with high accuracy. Then, referring to the data shown in FIG. 3, the magnet parameter estimation unit 70 calculates the temperature Tm of the permanent magnets 10a and 10b from the fundamental wave component Φa_1st of the magnetic flux density by the permanent magnet calculated by the fundamental wave magnetic flux calculation unit 61. It can be estimated with high accuracy. Thus, in the electric motor including a variable mechanism that changes the demagnetization state (magnetic flux density) of the low coercive force magnet portions 12a and 12b, whether the change in the induced voltage (magnetic flux density) is due to the change in the demagnetization state, It can be accurately determined whether it is due to a temperature change of the permanent magnets 10a and 10b. Therefore, even if the temperature Tm of the permanent magnets 10a and 10b and the magnetic flux density do not correspond one to one by the variable mechanism of the magnetic flux density, the temperature Tm of the permanent magnets 10a and 10b can be estimated with high accuracy. . That is, the variable amount by the temperature Tm of the permanent magnets 10a and 10b and the magnetic flux density variable mechanism can be estimated without measuring the variable amount by the magnetic flux density variable mechanism.

  Usually, the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density can be expressed by the equations (2) and (3) by the temperature Tm of the permanent magnets 10a and 10b and the variable amount V of the magnetic flux density. In the expressions (2) and (3), A to D are constants specific to the electric motor 100, respectively.

Φa_1st = A · Tm + B · V (2)
Φa_nst = C · Tm + D · V (3)

  By substituting the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density into the equations (2) and (3) and solving the simultaneous equations of the equations (2) and (3), the permanent magnet 10a The temperature Tm of 10b and the variable amount V of the magnetic flux density can be obtained respectively.

  FIG. 4 is a partial cross-sectional view showing a second example of the structure of the electric motor in the first embodiment of the present invention. The mechanism for varying the magnetic flux density in the present embodiment is not limited to that shown in FIG. An electric motor having another variable magnetic flux density mechanism will be described with reference to FIG.

  Compared with FIG. 2, the motor shown in FIG. 4 is such that the core 8 for short-circuiting the magnetic path can be freely inserted and removed in the gap between the adjacent permanent magnets 10a and 10b, and the permanent magnets 10a and 10b are high coercivity magnets. It is comprised only by the parts 11a and 11b, and the point which is not provided with the low coercive force magnet parts 12a and 12b of FIG. Other configurations are the same as those in FIG. 2, and illustration and description thereof are omitted.

  In the absence of the core 8, the gap between the adjacent permanent magnets 10a and 10b is filled with air. Since air is less likely to pass magnetic flux than the core 8, the magnetic flux generated by the permanent magnets 10 a and 10 b is less likely to be short-circuited in the rotor 4, and the magnetic flux density leading from the rotor 4 to the stator 2 is increased. On the other hand, when the core 8 is present, the magnetic flux generated by the permanent magnets 10a and 10b is easily short-circuited in the rotor 4, and the magnetic flux density leading from the rotor 4 to the stator 2 is reduced. As described above, the electric motor shown in FIG. 4 has a mechanism for changing the magnetic flux density by inserting and removing the core 8.

  With reference to FIG. 5, an example of data indicating the correlation between the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density and the temperature Tm of the permanent magnets 10a and 10b in the electric motor of FIG. 4 will be described. The graph of FIG. 5 is stored in the correlation data storage unit 71 as in FIG. The graph shown in FIG. 5 shows that the temperatures of the permanent magnets 10a and 10b are 60 ° C., 100 ° C. and 140 ° C. for the state where the core 8 is present (S) and the state where the core 8 is not present (A) in the electric motor of FIG. The magnetic flux density of the permanent magnets 10a and 10b when changed to ° C. is expressed for each order by Fourier series expansion. The horizontal axis indicates the order, and the vertical axis indicates the magnetic flux density. The horizontal axis shows only odd-order components of 1 (fundamental wave), 3, 5,. In FIG. 5, “S” indicates a state where the core 8 is present, and “A” indicates a state where the core 8 is not present.

  As shown in FIG. 5, the seventh harmonic component Φa_7st of the magnetic flux density is less dependent on the temperature of the permanent magnets 10 a and 10 b than the fundamental wave and other harmonic components, and the presence or absence of the core 8. High dependency. On the other hand, the fundamental wave component Φa_1st of the magnetic flux density is more dependent on the temperature of the permanent magnets 10a and 10b than the harmonic component Φa_nst. Therefore, for example, if the nth-order harmonic magnetic flux calculation unit 62 calculates the seventh-order harmonic component Φa_7st of the magnetic flux density, the presence or absence of the core 8 can be accurately estimated with reference to the data shown in FIG. . Referring to the data shown in FIG. 5, the temperature Tm of the permanent magnets 10a and 10b is accurately estimated from the fundamental wave component Φa_1st of the magnetic flux density by the permanent magnets 10a and 10b calculated by the fundamental wave magnetic flux calculation unit 61. be able to. Thus, in an electric motor capable of controlling the presence or absence of the core 8, whether the change in induced voltage (magnetic flux density) is due to the presence or absence of the core 8 or the temperature change of the permanent magnets 10a and 10b is accurately determined. Judgment can be made.

  FIG. 6A is a partial cross-sectional view showing a third example of the structure of the electric motor according to the first embodiment of the present invention, and FIG. 6B shows the rotating shaft 9 of the electric motor of FIG. FIG. Fig.6 (a) is sectional drawing along the AA cut surface of FIG.6 (b). The mechanism for varying the magnetic flux density in the present embodiment is not limited to that shown in FIGS. An example of an electric motor having another variable magnetic flux density mechanism will be described with reference to FIG.

  The electric motor shown in FIG. 6A and FIG. 6B has a double rotor structure including a first rotor 4a and a second rotor 4b arranged in the direction of the rotation shaft 9, and the first rotor 4a and the second rotor It has a variable mechanism that changes the magnetic flux density (induced voltage) of the entire rotor in accordance with the phase difference (skew) SK between the rotors 4b. The phase difference (skew) SK is an angle formed by the d-axis dAX1 of the first rotor 4a and the d-axis dAX2 of the second rotor 4b. Further, the permanent magnets 10a and 10b are composed only of the high coercive force magnet portions 11a and 11b, and do not include the low coercive force magnet portions 12a and 12b shown in FIG. Other configurations are the same as those in FIG. 2, and illustration and description thereof are omitted.

  With reference to FIG. 7 (a), the data of the correlation between the fundamental wave component and the nth harmonic component of the phase and the phase difference (skew) SK in the electric motors of FIG. 6 (a) and FIG. 6 (b). An example will be described.

  The graph shown in FIG. 7A is stored in the correlation data storage unit 71 as in FIG. 3 and FIG. The graph shown in FIG. 7A shows the case where there is a phase difference (skew) SK in the electric motors shown in FIGS. 6A and 6B (S) and the case where there is no phase difference (skew) SK (N). The phases when the temperatures of the permanent magnets 10a and 10b are respectively changed to 60 ° C., 100 ° C., and 140 ° C. are expressed for each order when Fourier series expansion is performed. The horizontal axis indicates the order, and the vertical axis indicates the phase. The horizontal axis shows only odd-order components of 1 (fundamental wave), 3, 5,. In FIG. 7A, “S” indicates a state where there is a phase difference (skew) SK, and “N” indicates a state where there is no phase difference (skew) SK.

  As shown in FIG. 7A, both the fundamental wave of the phase and the n-th harmonic do not depend on the temperature Tm of the permanent magnets 10a and 10b. The fundamental wave of the phase has no dependency on the presence or absence of the phase difference (skew) SK, but the third harmonic component and the fifth harmonic component of the phase have dependency on the presence or absence of the phase difference (skew) SK. Therefore, for example, if the nth-order harmonic magnetic flux calculation unit 62 calculates the third-order harmonic component or the fifth-order harmonic component of the phase, the phase difference (skew) is referred to with reference to the data shown in FIG. The presence or absence of SK can be accurately estimated. Then, referring to data representing the temperature dependence of the fundamental wave component Φa — 1st of the magnetic flux density as shown in FIG. 7B, the fundamental wave component Φa — 1st of the magnetic flux density calculated by the fundamental wave magnetic flux calculator 61 is permanently The temperature Tm of the magnets 10a and 10b can be estimated with high accuracy. Thus, in an electric motor capable of controlling the presence or absence of phase difference (skew) SK, whether the change in induced voltage (magnetic flux density) is due to the presence or absence of phase difference (skew) SK or the temperature of permanent magnets 10a and 10b. It is possible to accurately determine whether the change is caused.

  As shown in FIG. 2, FIG. 4, FIG. 6 (a) and FIG. 6 (b), the fundamental wave of the magnetic flux density (amplitude) or phase is used even in an electric motor having various types of magnetic flux density variable mechanisms. The component and the nth harmonic component change independently according to the temperature Tm of the permanent magnets 10a and 10b and the change in magnetic flux density due to the variable mechanism. The rate of change of the nth harmonic component of the magnetic flux density with respect to the change in the temperature Tm of the permanent magnets 10a and 10b is smaller than the rate of change of the fundamental wave component of the magnetic flux density with respect to the change in the temperature Tm of the permanent magnets 10a and 10b. Therefore, by using the fundamental wave component and the harmonic component of the magnetic flux density (amplitude) or phase, the change of the magnetic flux density due to the variable mechanism and the change of the magnetic flux density due to the temperature change of the permanent magnets 10a and 10b are separated. Can do.

  Here, the variable mechanism included in the electric motor 100 includes a magnetic flux density by the permanent magnets 10 a and 10 b, an induced voltage generated in the armature coil 6, and an inductance of the armature coil 6 according to the operating conditions of the electric motor 100. As long as at least one of them is changed. The correlation data storage unit 71 in FIG. 1B includes the fundamental wave component Φa_1st and the nth-order harmonic component of the magnetic flux density (amplitude) or phase shown in FIGS. 3, 5, 7A, and 7B. Data indicating the correlation between Φa_nst, the temperature of the permanent magnets 10a and 10b, and the motor constant described above is stored. The magnet parameter estimation unit 70 in FIG. 1B estimates the temperature Tm of the permanent magnets 10a and 10b and the above-described electric motor constant from the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density (amplitude) or phase. .

  In addition, instead of inputting the fundamental wave detection values id and iq converted into the dq coordinate system by the current vector control unit 32b into the fundamental wave magnetic flux calculation unit 61 of FIG. The removed feedback signal may be input. As a result, a signal from which noise has been removed from the detection values id and iq of the fundamental wave can be input to the fundamental wave magnetic flux calculator 61.

  Similarly, instead of inputting the detected values id_n and iq_n of the nth-order harmonics converted into the dq coordinate system by the current vector control unit 42b into the nth-order harmonic magnetic flux calculation unit 62 in FIG. A feedback signal from which the fundamental wave component and other high-frequency components are removed by 43 may be input. As a result, a signal from which noise has been removed from the detected values id_n and iq_n of the nth harmonic can be input to the nth harmonic magnetic flux calculator 62. The low-pass filter 33 and the low-pass filter 43 are an example of the “noise removal unit” in the present invention.

  Further, instead of the currents id, iq, id_n, iq_n flowing through the armature coil 6, voltages vd, vq, vd_n, vq_n applied to the armature coil 6, inductance of the armature coil 6, generated in the motor 100 Noise removal may be performed on the torque to be generated and the fundamental wave and the nth harmonic of the position of the rotor 4. In this case, noise-removed voltages vd, vq, vd_n, vq_n applied to the armature coil 6, inductance of the armature coil 6, torque generated in the motor 100, or a fundamental wave of the position of the rotor 4 and The fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density by the permanent magnets 10a and 10b can be calculated from the nth harmonic.

  Next, an example of a procedure for estimating the magnetic flux density and the temperature Tm of the permanent magnets 10a and 10b will be described for the electric motor 100 having a variable mechanism that changes the magnetic flux density by the permanent magnets 10a and 10b with reference to FIG. First, in step S101, the fundamental wave and the nth harmonic of the electric motor 100 are detected. An odd-order harmonic of 3 or more is detected as the n-order harmonic, but a value “n” is set according to the characteristics of the electric motor 100. Proceeding to step S102, the fundamental wave magnetic flux calculator 61 calculates the fundamental wave component of the magnetic flux density (amplitude) and phase by the permanent magnets 10a and 10b from the fundamental wave of the electric motor 100 according to the equation (1). Then, the n-order harmonic magnetic flux calculator 62 calculates the magnetic flux density (amplitude) and phase n-order harmonic components of the permanent magnets 10a and 10b from the n-order harmonic of the electric motor 100 according to the equation (1).

  Proceeding to step S103, the magnetic flux density distribution in the circumferential direction of the electric motor 100 is calculated from the magnetic flux density (amplitude), the fundamental wave component of the phase, and the nth harmonic component.

  With reference to FIG. 9, a method of calculating the magnetic flux density distribution in the circumferential direction of the electric motor 100 from the fundamental wave component and the nth harmonic component of the magnetic flux density (amplitude) and phase will be described. As shown in FIG. 9, when the magnetic flux density distribution of the permanent magnets 10a and 10b in the circumferential direction is uniform, the magnetic flux density distribution is a rectangular wave GA. The rectangular wave GA is represented by combining the fundamental wave component WB of the magnetic flux density and the odd-order harmonic components (third-order harmonic component W3, fifth-order harmonic component W5,...). Here, as shown in FIG. 2 (a), when the low coercive force magnet portions 12a and 12b located at the central portions of the permanent magnets 10a and 10b are demagnetized to 50%, the distribution of the magnetic flux density is a rectangular wave. It deforms from GA to a concave shape with a concave central part. By this deformation, the amplitude of the third harmonic component W3 becomes larger than the amplitudes of the fundamental wave component and other harmonic components. On the contrary, for example, when the low coercive force magnet portions 12a and 12b demagnetized to 50% are arranged in the peripheral portions of the permanent magnets 10a and 10b, or the temperature of the peripheral portion HT of the permanent magnets 10a and 10b is locally When the magnetic flux in the peripheral portion HT of the permanent magnets 10a and 10b is decreased due to the increase in the magnetic field, the distribution of the magnetic flux density is deformed from the rectangular wave GA to the convex shape GB protruding from the central portion as shown in FIG. To do. By this deformation, the amplitudes of the third harmonic component W3 and the fifth harmonic component W5 become smaller than the amplitude of the fundamental wave component. Thus, the distribution of the magnetic flux density in the circumferential direction of the electric motor 100 can be calculated from the magnetic flux density (amplitude), the fundamental wave component of the phase, and the nth harmonic component.

  Returning to FIG. 8, in step S104, the magnet parameter estimation unit 70 refers to the data stored in the correlation data storage unit 71, and determines the permanent magnet 10a from the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density. The amount of change in magnetic flux density due to the temperature Tm of 10b and the variable mechanism included in the electric motor 100 is estimated. Here, the “temperature Tm of the permanent magnets 10a and 10b” includes the temperature distribution of the permanent magnets 10a and 10b in the circumferential direction of the electric motor 100 in addition to the temperature of the permanent magnets 10a and 10b.

  As described above, according to the first embodiment of the present invention, the following operational effects can be obtained.

  The increase / decrease characteristic of the fundamental wave component Φa_1st of the magnetic flux density by the permanent magnets 10a and 10b is different from the increase / decrease characteristic of the nth-order harmonic component Φa_nst of the magnetic flux density. Therefore, by using the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density, even if the temperature Tm and the magnetic flux density of the permanent magnets 10a and 10b do not correspond one-to-one, the permanent magnet 10a, The temperature Tm or magnetic flux density of 10b can be estimated. Therefore, it is not necessary to make an excessively large heat-resistant margin for the permanent magnets 10a and 10b, and the manufacturing cost of the permanent magnets 10a and 10b can be reduced. In addition, torque accuracy, motor control efficiency, and current response are improved.

  Even when the motor 100 has a motor constant variable mechanism, the correlation between the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density, the temperature Tm of the permanent magnets 10a and 10b, and the motor constant is shown. By referring to the data, the temperature Tm and the motor constant of the permanent magnets 10a and 10b can be accurately estimated without measuring the amount of change in the motor constant by the variable mechanism.

  The fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density change independently according to the temperature Tm and the motor constant of the permanent magnets 10a and 10b. The rate of change of the nth-order harmonic component Φa_nst of the magnetic flux density with respect to changes in the temperature Tm of the permanent magnets 10a and 10b is smaller than the rate of change of the fundamental wave component Φa_1st of magnetic flux density with respect to changes in the temperature Tm of the permanent magnets 10a and 10b. Therefore, by using the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density, it is possible to separate the change of the magnetic flux density due to the variable mechanism and the change of the magnetic flux density due to the temperature change of the permanent magnets 10a and 10b. it can.

  The fundamental wave of the electric motor 100 is any one of the current flowing through the armature coil 6, the voltage applied to the armature coil 6, the inductance of the armature coil 6, the torque generated in the electric motor 100, and the position of the rotor 4. This is the fundamental wave. In addition, the nth harmonic of the motor 100 includes the current flowing through the armature coil 6, the voltage applied to the armature coil 6, the inductance of the armature coil 6, the torque generated in the motor 100, and the position of the rotor 4. Any one or more of the n-order harmonics. Using various parameters of the electric motor 100, the temperature Tm and the electric motor constant of the permanent magnets 10a and 10b can be estimated.

  The control device includes a current flowing through the armature coil 6, a voltage applied to the armature coil 6, an inductance of the armature coil 6, a torque generated in the electric motor 100, and a plurality of frequency components at the position of the rotor 4. Low pass filters 33 and 43 (noise removal units) for removing other frequency components excluding the fundamental wave and the nth harmonic are provided. Thereby, it is possible to accurately estimate the temperature Tm and the motor constant of the permanent magnets 10a and 10b using the fundamental wave and the nth harmonic from which noise has been removed.

(Second Embodiment)
In the flowchart of FIG. 8, the procedure for estimating the magnetic flux density and the temperature Tm of the permanent magnets 10a and 10b has been described for the electric motor 100 having a variable mechanism that changes the magnetic flux density of the permanent magnets 10a and 10b as the motor constant. In 2nd Embodiment, with reference to FIG. 10, about the motor 100 which has a variable mechanism which changes the induced voltage which arises in the armature coil 6 as a motor constant, the procedure which estimates the temperature of induced voltage and permanent magnet 10a, 10b An example will be described. The processing content in step S105 in FIG. 10 is different from that in step S104 in FIG. 8, and the processing content in the other steps is the same as that in FIG.

  In step S105 of FIG. 10, the magnet parameter estimation unit 70 refers to the data stored in the correlation data storage unit 71 and calculates the temperature of the permanent magnets 10a and 10b from the fundamental wave component and the nth harmonic component of the magnetic flux density. , And the amount of change in the induced voltage due to the variable mechanism included in the electric motor 100 is estimated. In the second embodiment, the correlation data storage unit 71 stores data indicating the correlation between the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density, the temperature Tm of the permanent magnets 10a and 10b, and the variable amount of the induced voltage. It is remembered.

(Third embodiment)
In the third embodiment, referring to FIG. 11, an example of a procedure for estimating the inductance and the temperature Tm of the permanent magnets 10a and 10b for the motor 100 having a variable mechanism that changes the inductance of the armature coil 6 as the motor constant. Will be explained. The processing content in step S106 in FIG. 11 is different from that in step S104 in FIG. 8, and the processing content in the other steps is the same as in FIG.

  In step S106 of FIG. 11, the magnet parameter estimation unit 70 refers to the data stored in the correlation data storage unit 71 and calculates the temperature of the permanent magnets 10a and 10b from the fundamental wave component and the nth harmonic component of the magnetic flux density. And the amount of change in inductance by the variable mechanism included in the electric motor 100 is estimated. In the third embodiment, the correlation data storage unit 71 stores data indicating the correlation between the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density, the temperature Tm of the permanent magnets 10a and 10b, and the variable amount of inductance. Has been.

(Fourth embodiment)
In the fourth embodiment, with reference to FIG. 12, the fundamental waves id and iq of the current flowing through the armature coil 6 and the nth harmonics id_n and iq_n are detected as the fundamental wave and the nth harmonic of the electric motor 100. The temperature estimation method will be described.

  In step S201, the fundamental waves id and iq of the current flowing through the armature coil 6 are detected as the fundamental wave of the motor 100, and the nth harmonic id_n of the current flowing through the armature coil 6 as the nth harmonic of the motor 100 is detected. , Iq_n. Then, in step S202, the fundamental wave magnetic flux calculator 61 calculates the fundamental wave component Φa_1st of the magnetic flux density from the fundamental waves id and iq of the current using the equation (1). The nth-order harmonic magnetic flux calculation unit 62 calculates the nth-order harmonic component Φa_nst of the magnetic flux density from the nth-order harmonics id_n and iq_n of the current using the equation (1). Step S203 is the same as step S103 in FIG. In step S204, the magnet parameter estimation unit 70 refers to the data stored in the correlation data storage unit 71 and determines the temperature of the permanent magnets 10a and 10b from the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density. The amount of change in the motor constant by Tm and the variable mechanism included in the motor 100 is estimated.

(Fifth embodiment)
In the fifth embodiment, as shown in FIG. 13, in step S205, the fundamental waves vd and vq of the voltage applied to the armature coil 6 and the n-order harmonics vd_n are used as the fundamental wave and the n-order harmonics of the electric motor 100. , Vq_n are detected. In step S202, the fundamental wave magnetic flux calculation unit 61 calculates the fundamental wave component Φa_1st of the magnetic flux density from the fundamental waves vd and vq of the voltage using the equation (1). The nth-order harmonic magnetic flux calculation unit 62 calculates the nth-order harmonic component Φa_nst of the magnetic flux density from the nth-order harmonics vd_n and vq_n of the voltage using the equation (1). Steps S203 and S204 are the same as those in FIG.

(Sixth embodiment)
In the sixth embodiment, as shown in FIG. 14, in step S206, the fundamental wave and the nth harmonic of the inductance of the armature coil 6 are detected as the fundamental wave and the nth harmonic of the motor 100. The fundamental wave of the inductance of the armature coil 6 is obtained from the fundamental waves vd and vq of the voltage applied to the armature coil 6 and the fundamental waves id and iq of the current flowing through the armature coil 6 by using a known calculation method. Can be calculated. The same applies to the nth harmonic of the inductance of the armature coil 6.

  In step S202, the fundamental wave magnetic flux calculator 61 calculates the fundamental wave component Φa_1st of the magnetic flux density from the fundamental wave of the inductance using a known calculation method. The nth-order harmonic magnetic flux calculation unit 62 calculates the nth-order harmonic component Φa_nst of the magnetic flux density from the nth-order harmonic of the inductance using a known calculation method. Steps S203 and S204 are the same as those in FIG.

(Seventh embodiment)
In the seventh embodiment, as shown in FIG. 15, in step S207, the fundamental wave and the nth harmonic of the torque generated in the motor 100 are detected as the fundamental wave and the nth harmonic of the motor 100. The fundamental wave of the torque generated in the electric motor 100 is obtained by using a known calculation method from the fundamental waves vd and vq of the voltage applied to the armature coil 6 and the fundamental waves id and iq of the current flowing through the armature coil 6. Can be calculated. The same applies to the nth harmonic of the torque generated in the electric motor 100.

  In step S202, the fundamental wave magnetic flux calculator 61 calculates the fundamental wave component Φa_1st of the magnetic flux density from the fundamental wave of the torque using a known calculation method. The n-order harmonic magnetic flux calculation unit 62 calculates the n-order harmonic component Φa_nst of the magnetic flux density from the n-order harmonic of the torque using a known calculation method. Steps S203 and S204 are the same as those in FIG.

(Eighth embodiment)
In the eighth embodiment, as shown in FIG. 16, in step S208, the fundamental wave and the nth harmonic of the position (rotation angle) of the rotor 4 are detected as the fundamental wave and the nth harmonic of the electric motor 100. The fundamental wave at the position of the rotor 4 is calculated from the fundamental waves vd and vq of the voltage applied to the armature coil 6 and the fundamental waves id and iq of the current flowing through the armature coil 6 using a known calculation method. be able to. The same applies to the nth harmonic at the position of the rotor 4.

  In step S202, the nth-order harmonic magnetic flux calculation unit 62 calculates the nth-order harmonic component Φa_nst of the magnetic flux density from the nth-order harmonic at the position using a known calculation method. Steps S203 and S204 are the same as those in FIG.

(Ninth embodiment)
12 to 16, as the fundamental wave and the nth harmonic of the electric motor 100, the current flowing through the armature coil 6, the voltage applied to the armature coil 6, the inductance of the armature coil 6, and the electric motor 100 are generated. In the above description, the fundamental wave and the nth harmonic of any one of the torque and the position of the rotor 4 are detected. In the ninth embodiment, any one of the current flowing through the armature coil 6, the voltage applied to the armature coil 6, the inductance of the armature coil 6, the torque generated in the electric motor 100, and the position of the rotor 4 is selected. A case where the fundamental wave and the n-th harmonic of two or more electric motors 100 are detected will be described.

  In the ninth embodiment, as shown in FIG. 17, in step S <b> 209, the basic current and the voltage applied to the armature coil 6 as the fundamental wave and the nth harmonic of the motor 100 are applied to the armature coil 6. Waves id, iq, vd, vq and n-th harmonics id_n, iq_n, vd_n, vq_n are detected.

  In step S202, the fundamental wave magnetic flux calculator 61 calculates a fundamental wave component Φa_1st of the magnetic flux density from the fundamental waves id, iq, vd, and vq of the current and voltage according to the equation (1). The nth-order harmonic magnetic flux calculation unit 62 calculates the nth-order harmonic component Φa_nst of the magnetic flux density from the nth-order harmonics id_n, iq_n, vd_n, and vq_n of current and voltage according to the equation (1). Steps S203 and S204 are the same as those in FIG.

(10th Embodiment)
In the tenth embodiment, referring to FIG. 18, the motor constant and the position estimation error of the rotor 4 are combined with the detected value of the current flowing through the armature coil 6 and the voltage applied to the armature coil 6. An example of the procedure for estimating the magnet temperature will be described.

  First, in step S210, the current flowing through the armature coil 6, the voltage applied to the armature coil 6, and the position of the rotor 4 are detected. The position of the rotor 4 is the rotation angle θ of the rotor 4 detected by the position sensor 104. In step S211, the position of the rotor 4 is estimated using a known method using the current flowing through the armature coil 6 and the voltage applied to the armature coil 6. In step S212, a position estimation error of the rotor 4 is detected from the estimated value of the position of the rotor 4 and the detected value of the position of the rotor 4. Further, the same processing as in steps S209, S202, and S203 shown in FIG. 17 is performed. In step S204, the magnet parameter estimation unit 70 considers the position estimation error of the rotor 4 and calculates the temperature Tm of the permanent magnets 10a and 10b and the electric motor from the fundamental wave component Φa_1st and the nth harmonic component Φa_nst of the magnetic flux density. The change amount of the motor constant by the variable mechanism included in 100 is estimated. By considering the position estimation error of the rotor 4, the estimation accuracy of the change amount of the temperature Tm and the motor constant is improved.

(Eleventh embodiment)
In the first to tenth embodiments, the motor 100 including the variable mechanism that changes the motor constant according to the operating condition of the motor 100 has been described. However, in the eleventh embodiment, the variable mechanism that changes the motor constant is not provided. An example in which the present invention is applied to the electric motor 100 will be described.

  As described with reference to FIG. 9, in the electric motor 100 that does not include the magnetic flux density variable mechanism, the magnetic flux density distribution by the permanent magnets 10a and 10b is a rectangular wave GA. However, for example, when the temperature of the peripheral portion HT of the permanent magnets 10a and 10b is locally increased, the magnetic flux density of the peripheral portion HT is smaller than that of the central portion, and the distribution of the magnetic flux density is changed from the rectangular wave GA to the central portion. Is deformed into a protruding shape GB. As described above, the distribution of the magnetic flux density can be represented by a combination of the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density. When the rectangular wave GA is deformed to the convex shape GB, the third harmonic component Φa_3st becomes smaller than the fundamental wave component Φa_1st and the fifth harmonic component Φa_5st. As described above, the temperature distribution of the permanent magnets 10a and 10b can be obtained from the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density even in the electric motor 100 that does not include the variable magnetic flux density mechanism.

  FIG. 19 is a flowchart illustrating an example of a procedure for estimating the temperature Tm of the permanent magnets 10a and 10b in the electric motor 100 that does not include the magnetic flux density variable mechanism. Since the electric motor 100 is not provided with a magnetic flux density variable mechanism, step S103 in FIG. 8 for obtaining the magnetic flux density distribution is omitted. The other steps S101, S102, and S104 are the same as those in FIG.

(Twelfth embodiment)
When the rotor 4 is not rotating, or when the rotational speed of the rotor 4 is in a low rotation state, no induced voltage is generated in the electric motor 100 or the induced voltage is weak. Therefore, the fundamental wave components Φa_1st and n of the magnetic flux density The second harmonic component Φa_nst cannot be detected, or the detection accuracy thereof decreases, and consequently the temperature estimation accuracy of the permanent magnets 10a and 10b also decreases. Therefore, in the twelfth embodiment, when the rotation speed of the rotor 4 is lower than a predetermined threshold value, a high frequency that does not contribute to the rotation of the rotor 4 is superimposed on the armature coil 6 and the superimposed high frequency fundamental wave. A method for estimating the motor constant and the temperature Tm of the permanent magnets 10a and 10b using the nth harmonic will be described with reference to FIG.

  In step S301, the position sensor 104 (rotational speed determination unit) determines whether the rotational speed (electrical angular velocity ω) of the rotor 4 is lower than a predetermined threshold value. When it is determined that the rotational speed of the rotor 4 is lower than the predetermined threshold (YES in S301), a high frequency voltage is superimposed on the armature coil 6 in step S302.

  In step S303, the current vector control unit 32b in FIG. 1A detects the fundamental waves id and iq of the high-frequency current flowing through the armature coil 6, and the current vector control unit 42b flows through the armature coil 6. N-order harmonics id_n and iq_n of the high-frequency current are detected.

  Proceeding to step S304, the fundamental wave magnetic flux calculator 61 calculates the fundamental wave components of the magnetic flux density (amplitude) and phase by the permanent magnets 10a and 10b from the fundamental waves id and iq of the high frequency current according to the equation (1). Then, the n-order harmonic magnetic flux calculation unit 62 calculates the magnetic flux density (amplitude) and phase n-order harmonic components of the permanent magnets 10a and 10b from the n-order harmonics id_n and iq_n of the high-frequency current in accordance with the equation (1). To do.

  Proceeding to step S305, the distribution of the magnetic flux density in the circumferential direction of the electric motor 100 is calculated from the magnetic flux density (amplitude), the fundamental wave component of the phase, and the nth harmonic component. Then, similarly to step S104 in FIG. 8, the magnet parameter estimation unit 70 refers to the data stored in the correlation data storage unit 71 and determines the permanent magnet from the fundamental wave component and the nth harmonic component of the magnetic flux density. The amount of change in the induced voltage due to the temperature of 10a, 10b and the variable mechanism provided in the electric motor 100 is estimated.

  On the other hand, when it is determined that the rotational speed of the rotor 4 is higher than the predetermined threshold (NO in S301), steps S101 to S014 are performed according to the flowchart shown in FIG.

  Instead of superimposing a high frequency voltage on the armature coil 6, a high frequency current may be superimposed on the armature coil 6, as shown in FIG. In this case, in step S303, the fundamental waves vd and vq of the high frequency voltage applied to the armature coil 6 and the nth harmonics vd_n and vq_n of the high frequency voltage applied to the armature coil 6 are detected.

  Then, in step S304, the fundamental wave magnetic flux calculator 61 calculates the fundamental wave components of the magnetic flux density (amplitude) and phase by the permanent magnets 10a and 10b from the fundamental waves vd and vq of the high frequency voltage according to the equation (1). . Then, the n-order harmonic magnetic flux calculation unit 62 calculates the magnetic flux density (amplitude) and the n-order harmonic component of the phase by the permanent magnets 10a and 10b from the n-order harmonics vd_n and vq_n of the high-frequency voltage according to the equation (1). To do.

  Thus, by superimposing a high-frequency voltage or high-frequency current that does not contribute to the rotation of the rotor 4 on the armature coil 6, an induced voltage that does not contribute to the rotation of the rotor 4 is generated in the electric motor 100. Using this induced voltage, the fundamental wave component Φa_1st and the nth-order harmonic component Φa_nst of the magnetic flux density are detected, and the temperature of the permanent magnets 10a and 10b and the amount of change in the induced voltage due to the variable mechanism provided in the electric motor 100 are accurately estimated. can do.

  Further, when the rotor 4 is not rotating or when the rotational speed of the rotor 4 is in a low rotation state, a high-frequency voltage or a high-frequency current that does not contribute to the rotation of the rotor 4 is superimposed on the armature coil 6. When the rotational speed of 4 is in a high rotational state, no high frequency voltage or high frequency current is superimposed. Thereby, a high frequency voltage or a high frequency current can be efficiently superimposed.

  Although the embodiments of the present invention have been described as described above, it should not be understood that the descriptions and drawings constituting a part of this disclosure limit the present invention. From this disclosure, various alternative embodiments, examples and operational techniques will be apparent to those skilled in the art.

DESCRIPTION OF SYMBOLS 2 ... Stator 4 ... Rotor 6 ... Armature coil 10a, 10b ... Permanent magnet 11a, 11b ... High coercive force magnet part 12a, 12b ... Low coercive force magnet part 31, 41 ... Current controller 32a, 32b, 42a, 42b ... Current Vector control unit 33, 43 ... low pass filter 61 ... fundamental wave magnetic flux calculation unit 62 ... nth harmonic magnetic flux calculation unit 70 ... magnet parameter estimation unit 71 ... correlation data storage unit 100 ... electric motor 103 ... current sensor 104 ... position sensor Φa_1st ... Fundamental wave component of magnetic flux density Φa_nst: nth-order harmonic component of magnetic flux density θ: rotation angle (rotor position)
id, iq ... fundamental wave (current)
id_n, iq_n ... nth harmonic (current)
vd, vq ... fundamental wave (voltage)
vd_n, vq_n ... nth harmonic (voltage)

Claims (8)

A control device that controls an electric motor including a rotor having a permanent magnet and a stator having an armature coil by using a fundamental wave of a driving frequency and n (n is a natural number of 2 or more) order harmonics,
A fundamental wave detector for detecting the fundamental wave of the electric motor;
An nth harmonic detection unit for detecting the nth harmonic of the electric motor;
A fundamental magnetic flux calculation unit for obtaining a fundamental wave component of magnetic flux density by the permanent magnet from the fundamental wave of the electric motor;
An n-order harmonic magnetic flux calculation unit for obtaining an n-order harmonic component of the magnetic flux density by the permanent magnet from the n-order harmonic of the electric motor;
A correlation data storage unit that stores data indicating the correlation between the fundamental wave component and the nth harmonic component of the magnetic flux density and the temperature of the permanent magnet;
A magnet parameter estimation unit that estimates the temperature or magnetic flux density of the permanent magnet from the fundamental wave component and the nth harmonic component of the magnetic flux density with reference to data stored in the correlation data storage unit,
An electric motor control device. The electric motor changes at least one of an electric motor constant including a magnetic flux density by the permanent magnet, an induced voltage generated in the armature coil, and an inductance of the armature coil according to an operating condition of the electric motor. An electric motor,
The correlation data storage unit stores data indicating a correlation between the fundamental wave component and the nth harmonic component of the magnetic flux density, the temperature of the permanent magnet, and the motor constant,
The magnet parameter estimation unit estimates the temperature of the permanent magnet and the motor constant from the fundamental wave component and the nth harmonic component of the magnetic flux density.
The motor control device according to claim 1. The fundamental wave component and the nth harmonic component of the magnetic flux density change independently according to the temperature of the permanent magnet and the motor constant,
The rate of change of the nth-order harmonic component of the magnetic flux density with respect to a change in the temperature of the permanent magnet is smaller than the rate of change of the fundamental wave component of the magnetic flux density with respect to a change in the temperature of the permanent magnet. Electric motor control device. The fundamental wave of the motor is any one of a current flowing through the armature coil, a voltage applied to the armature coil, an inductance of the armature coil, a torque generated in the motor, and a position of the rotor. One or more fundamental waves,
The n-th harmonic of the motor is a current flowing through the armature coil, a voltage applied to the armature coil, an inductance of the armature coil, a torque generated in the motor, and a position of the rotor. Any one or more of the n th harmonics,
The motor control device according to claim 1.   From the current flowing through the armature coil, the voltage applied to the armature coil, the inductance of the armature coil, the torque generated in the motor, and the plurality of frequency components of the position of the rotor, the fundamental wave and The motor control device according to claim 4, further comprising a noise removing unit that removes other frequency components excluding the n-th harmonic. The fundamental wave detection unit detects the fundamental wave of high frequency that is superimposed on the armature coil and does not contribute to the rotation of the rotor,
The nth harmonic detection unit detects the high frequency nth harmonic,
The fundamental magnetic flux calculator calculates a fundamental wave component of magnetic flux density by the permanent magnet from the fundamental wave of the high frequency,
The n-order harmonic magnetic flux calculation unit obtains an n-order harmonic component of the magnetic flux density by the permanent magnet from the high-frequency n-order harmonic.
The motor control device according to any one of claims 1 to 5. A rotational speed determination unit that determines whether the rotational speed of the rotor is lower than a predetermined value;
A high frequency superimposing unit that superimposes the high frequency on the armature coil when the rotational speed determining unit determines that the rotational speed of the rotor is lower than a predetermined value;
The motor control device according to claim 6, further comprising: A control method for controlling an electric motor including a rotor having a permanent magnet and a stator having an armature coil by using a fundamental wave of a driving frequency and n (n is a natural number of 2 or more) order harmonics,
Detecting the fundamental wave of the motor;
Detecting the nth harmonic of the motor;
Obtain the fundamental wave component of the magnetic flux density by the permanent magnet from the fundamental wave of the electric motor,
Obtaining the nth harmonic component of the magnetic flux density by the permanent magnet from the nth harmonic of the electric motor,
With reference to data indicating the correlation between the fundamental wave component and nth harmonic component of the magnetic flux density and the temperature of the permanent magnet, the temperature of the permanent magnet is obtained from the fundamental wave component and the nth harmonic component of the magnetic flux density. Or estimating the magnetic flux density,
Electric motor control method.

Images