Method and apparatus for controlling a synchronous motor

[0002] This application is a continuation of PCT Patent Application PCT/JP99/04461, filed Aug. 20, 1999, which was published in Japanese.

[0003] The present invention in general relates to an apparatus and method for controlling a synchronous motor. More particularly, this invention relates to an apparatus and method for controlling a synchronous motor having a magnet is buried in a rotor.

[0004] Various control methods related to a synchronous motor in which a permanent magnet is attached to the surface of the rotor (hereinbelow, called an SPM motor) and a synchronous motor in which a permanent magnet is buried in the rotor (hereinbelow, called an IPM motor), for driving the motor without providing a sensor for sensing a rotation position of the rotor are conventionally provided. For example, in (a) Japanese Patent Office Application Laid-Open No. 8-308286, (b) Japanese Patent Office Application Laid-Open No. 9-191698, and the like, a magnetic pole position is estimated on the basis of an error between an estimated current and an actual current, and a vector control of the synchronous motor is performed by using the result of estimation. Furthermore, (c) The Transactions of The Institute of Electrical Engineers of Japan, Vol. 111-D, No. 8, 1991, pp. 639-644 discloses a simple and easy voltage instruction operating method and a simple and easy primary angular frequency setting method in an SPM motor.

[0005] The method of (c) will be described in detail hereinbelow. It is additionally written that the γ-δ axis of (c) is opposite to that used in the present invention (the γ axis of (c) corresponds to the δ axis of the invention, and the δ axis of (c) corresponds to the γ axis of the invention).

[0006] FIG. 6 is a diagram showing the configuration of a conventional control system disclosed in (c).

[0007] synchronous motor controlling apparatus comprises following components. That is, an SPM motor 1, and an inverter 2 which applies a PWM modulated sine wave voltage to the SPM motor. A PWM control circuit 3 generates an ON/OFF signal which has been subjected to PWM to a switch in the inverter 2. A current detector 4 detects current flowing from the inverter 2 to the SPM motor 1. A current converter 5 converts the value of the current detected by the current detector 4 to a γ-δ axis assumed on the rotor. A microprocessor 6 receives the value of the current detected by the current detector 5, calculates a voltage instruction and a primary angular frequency as described in detail below, coordinate converts the voltage, and outputs the resultant to the PWM control circuit 3.

[0008] The microprocessor 6 calculates the voltage instruction based on the following equations (1) and (2).

[0009] Vγ=R1·I_γ+w1·λ1δ−w1·λδest  (1)

[0010] Vδ=R1·Iδ−Kδ·λδest  (2)

[0011] In these equations, R1 denotes primary resistance, w1 is a primary angular frequency, Kδ is a gain (constant), and λ1δ is a δ-axis component of a primary flux linkage. λδest is expressed by the equation (3).

[0012] λδest=L1·(Iδ−Iγ²/2I₀+I₀−Iδ*)  (3)

[0013] In this equation, L1 denotes an inductance of a primary winding, and I₀is obtained by dividing a flux linkage Λ₀by a magnet by L1. That is, I₀=Λ₀/L1. Iδ* is obtained by multiplying a primary flux linkage instruction Λδ* by L1. That is, Iδ*=Λδ*/L1.

[0014] The primary angular frequency w1 can be calculated using the following equation (4).

[0015] w1=w1*−Km·Iγ  (4)

[0016] In this equation, w1* is an angular velocity instruction and Km is a constant of proportionality.

[0017] In the conventional methods (a) and (b), however, the velocity is estimated and the vector control is performed. Consequently, gain adjustment of a velocity control system according to the inertia of a load connected to a motor shaft is necessary, and the computation is relatively complicated. In the conventional method (c), the gain setting due to the inertia of the load is unnecessary and the computation amount is small. However, the method (c) relates to the system of the SPM motor and cannot be applied to an IPM motor having different inductances Ld and Lq on a d-q axis fixed to a stator axis and a different torque generating method.

[0018] It is an object of the invention to obtain a control method which can be applied to an IPM motor, in which number of calculations is small, and which does not need gain adjustment according to the inertia of a load.

[0019] In order to perform a smooth start without a sensor, accurate primary resistance is necessary, so that another object of the invention is to realize a smooth start by obtaining accurate primary resistance in a short time at the time of starting.

[0020] FIG. 1 is a block diagram showing a synchronous motor controlling apparatus according to a first embodiment of the invention.

[0021] FIG. 2 is a block diagram showing a synchronous motor controlling apparatus according to a second embodiment of the invention.

[0022] FIG. 3 is a flowchart at the time of the start of a synchronous motor controlling apparatus according to a third embodiment of the invention.

[0023] FIG. 4 is a block diagram showing a resistance estimating method of the synchronous motor controlling apparatus in FIG. 3.

[0024] FIG. 5 is a block diagram showing another example of the resistance estimating method of the synchronous motor controlling apparatus in FIG. 3.

[0025] FIG. 6 is a block diagram showing a conventional synchronous motor controlling apparatus.

[0026] Preferred embodiments of the invention will be described as follows.

[0027] First Embodiment

[0028] FIG. 1 is a block diagram showing a synchronous motor controlling apparatus according to a first embodiment of the invention.

[0029] This synchronous motor controlling apparatus comprises following components. That is, an IPM motor 10 and an inverter circuit 11 which receives three-phase voltage instructions Vu, Vv, and Vw, converts the voltage instructions into PWM and outputs a PWM voltage through a switch. A current detector 12 detects a current flowing from the inverter circuit 11 to the IPM motor 10. A current coordinate transformer 13 transforms the values of the currents Iu and Iv detected by the current detector 12 onto a γ-δ axis assumed on a rotor to obtain currents Iγ and Iδ. A correction term Icmp computing unit 14 which calculates a correction term Icmp using the currents Iγ and Iδ. A Vγ and Vδ computing unit 15 which calculates voltage instructions Vγ and Vδ. A voltage coordinate transformer 16 which transforms the voltage instructions obtained by the Vγ and Vδ computing unit 15 into the three-phase voltage instructions Vu, Vv, and Vw which are supplied to the inverter circuit 11. A torque computing unit 17 which calculates IPM motor generation torque using the currents Iγ and Iδ. A speed stabilizing term computing unit 18 which calculates a speed stabilizing term W_Tfrom the torque obtained by the torque computing unit 17. A subtracting circuit 19 which subtracts the speed stabilizing term W_Tobtained by the speed stabilizing term computing unit 18 from the instructed speed W1* and obtains the primary angular frequency W1. An integrator 20 integrates the primary angular frequency W1 to obtain a phase θ which is used for coordination transformation in the current coordinate transformer 13 and the voltage coordinate transformer 16.

[0030] The components in FIG. 1 will now be described specifically. First, a voltage equation and a torque equation of the IPM motor 10 are as shown by equations (5) and (6) on the d-q axis fixed on the rotor axis. $\begin{array}{cc}\left[\begin{array}{c}\mathrm{Vd}\\ \mathrm{Vq}\end{array}\right]=\left[\begin{array}{cc}R+\mathrm{pLd}& -\omega \mathrm{Lq}\\ \omega \mathrm{Ld}& R+\mathrm{pLq}\end{array}\right]\left[\begin{array}{c}\mathrm{Id}\\ \mathrm{Iq}\end{array}\right]+\left[\begin{array}{c}0\\ \omega {\varphi }_f\end{array}\right]& \left(5\right)\end{array}$  Tm=Pm·φ_f·Iq+Pm·(Ld−Lq)·Id·Iq  (6)

[0031] In these equations, Vd and Vq are a d-axis component and a q-axis component of a primary voltage, respectively, R is primary resistance of the IPM motor, ω is a rotor rotation electric angular velocity, pm is the number of pairs of poles, f is a flux linkage by a magnet, Id and Iq are d-axis and q-axis currents, Ld and Lq are d-axis and q-axis inductances, and p is a differential operator d/dt.

[0032] In a stationary state, by setting the differential operator p in the equation (5) to zero, the equation (7) is obtained. $\begin{array}{cc}\left[\begin{array}{c}\mathrm{Vd}\\ \mathrm{Vq}\end{array}\right]=\left[\begin{array}{cc}R& -\omega 1\mathrm{Lq}\\ \mathrm{\omega 1}\mathrm{Ld}& R\end{array}\right]\left[\begin{array}{c}\mathrm{Id}\\ \mathrm{Iq}\end{array}\right]+\left[\begin{array}{c}0\\ \mathrm{\omega 1}·{\varphi }_f\end{array}\right]& \left(7\right)\end{array}$

[0033] Although the d-q axis is fixed on the rotor axis, the position of the rotor is not known in the case where a sensor is not provided. The γ-δ axis assumed on the stator is usually deviated from the d-q axis. In the case of performing the control in accordance with the invention, the deviation is a little, and the voltage equation in the stationary state on the γ-δ axis is approximated as shown by the following equation (8). $\begin{array}{cc}\left[\begin{array}{c}{V}_{\gamma }\\ V_{\delta }\end{array}\right]=\left[\begin{array}{cc}R& -\omega 1\mathrm{Lq}\\ \mathrm{\omega 1}\mathrm{Ld}& R\end{array}\right]\left[\begin{array}{c}{I}_{\gamma }\\ I_{\delta }\end{array}\right]+\left[\begin{array}{c}0\\ \mathrm{\omega 1}·{\varphi }_f\end{array}\right]& \left(8\right)\end{array}$

[0034] The correction term Icmp computing unit 14 calculates the correction term Icmp based on the following equation (9).

[0035] Icmp=Ix*−Iγ+Kcmp·Iδ  (9)

[0036] In equation (9), Ix* is a correction current instruction and Kcmp is a correction gain.

[0037] The Vγ and Vδ computing unit 15 multiplies the correction term Icmp obtained by the equation (9) by a proper gain and adds the resultant to the equation (8), that is, calculates as shown by the equation (10). $\begin{array}{cc}\left[\begin{array}{c}{V}_{\gamma }\\ V_{\delta }\end{array}\right]=\left[\begin{array}{cc}R& -\omega 1\mathrm{Lq}\\ \mathrm{\omega 1}\mathrm{Ld}& \end{array}\right]\left[\begin{array}{c}{I}_{\gamma }\\ I_{\delta }\end{array}\right]+\left[\begin{array}{c}0\\ \mathrm{\omega 1}·{\varphi }_f\end{array}\right]+\left[\begin{array}{c}\mathrm{Kgan}\\ \mathrm{Kdel}\end{array}\right]·\mathrm{Icmp}& \left(10\right)\end{array}$

[0038] With such a configuration, a voltage steadily applied when the currents Iγ and Iδ are passed is positively added and the correction term Icmp is added, thereby assuring stability of the voltage control also in an IPM motor.

[0039] The operations of the torque computing unit 17, speed stabilizing term computing unit 18, and subtracting circuit 19 will now be described. As shown by the example of the conventional technique, in the SPM motor, the speed can be stabilized by the equation (4). The stabilization of speed denotes prevention of loss of synchronization.

[0040] In the IPM motor, the torque is expressed by the equation (6). A reluctance torque expressed by the right side second term of the equation (6), which does not occur in the SPM motor occurs.

[0041] The torque computing unit 17 executes calculation as shown by the following equation (11).

[0042] T=Pm·φf·iδ+Pm·(Ld−Lq)·Iγ·Iδ  (11)

[0043] The speed stabilizing term computing unit 18 and the subtracting circuit 19 obtain the primary angular frequency w1 by using the torque T in a manner similar to the equation (4) shown in the example of the conventional technique, that is, as shown by the equations (12) and (13).

[0044] w_T=T·Km  (12)

[0045] w1=w1*−w_T  (13)

[0046] In a manner similar to the reference literature (c) described in the conventional technique, operations are made so that w_Tis steadily zero. In such a manner, the speed is stabilized and the control is performed steadily in a state where a speed deviation is zero.

[0047] According to the foregoing embodiment, the stable control can be performed by simple calculations such as the equations (9) and (10). Since the control is executed without estimating speed, the gain does not have to be adjusted according to the inertia of the load. Further, as a result of operating point analysis and actual verification, a high-efficiency control effectively using the reluctance torque can be realized by the control method.

[0048] Second Embodiment

[0049] FIG. 2 is a block diagram showing a synchronous motor controlling apparatus according to a second embodiment of the invention.

[0050] In this embodiment, the Vγ and Vδ computing unit 15 in FIG. 1 is replaced with a Vγ and Vδ computing unit 21. The components that perform the same or similar functions as those in the first embodiment have been provided with the same legends and, to avoid repetition of explanation, they will not be described.

[0051] When performing the control over a power running/regenerating range, the Vγ and Vδ computing unit 15 described in the first embodiment expressed by the equation (10) is not always stable in various IPM motors. By switching an Iδ component exerting an influence on Vγ and an Iγ component exerting an influence on Vδ, what is called an interference term voltage, so as to be considered or not according to the sign of Iδ, a stable control can be performed.

[0052] That is, when Iδ is positive, the following equation (14) is used. When Iδ is negative, the equation (10) is used. $\begin{array}{cc}\left[\begin{array}{c}{V}_{\gamma }\\ V_{\delta }\end{array}\right]=\left[\begin{array}{cc}R& 0\\ 0& R\end{array}\right]\left[\begin{array}{c}{I}_{\gamma }\\ I_{\delta }\end{array}\right]+\left[\begin{array}{c}0\\ \mathrm{\omega 1}·{\varphi }_f\end{array}\right]+\left[\begin{array}{c}\mathrm{Kgan}\\ \mathrm{Kdel}\end{array}\right]·\mathrm{icmp}& \left(14\right)\end{array}$

[0053] Further, the following was also confirmed by actual verification. When Iδ is negative, by setting the constant Kcmp of the correction term icmp computing unit 14 to zero or adjusting the interference term of Vδ, the torque characteristic can be improved by low speed regeneration. When Iδ changes from positive to negative, by gradually operating Kcmp and the interference term, the switching can be made without fluctuating speed at the time of switching.

[0054] According to the embodiment as described above, even an IPM motor which does not always have good characteristics in various speeds and torques in the control system described in the example of the conventional technique can have good torque characteristics.

[0055] In addition, the degree of freedom in designing is very high in an IPM motor. Even when stable controllability cannot be obtained in various speeds and torques, by operating the Ix* and Kcmp in the equation (9) and Kgan, Kdel, and the interference term in the equation (10), the stability can be assured and the excellent torque characteristics can be obtained.

[0056] Third Embodiment

[0057] FIG. 3 shows a flowchart at the time of start of a synchronous motor controlling apparatus according to a third embodiment of the invention. At step 1, an operation start signal is supplied. At step 2, an initial magnetic pole position at the time of stop is detected. Various methods are proposed as methods of detecting the initial magnetic pole position at the time of stop, such as “Initial Rotor Position Estimation of Sensorless Salient-Pole Brushless DC motor” (The Transactions of The Institute of Electrical Engineers of Japan, Vol. 116-D, No. 7, 1996) and “Evaluation of Estimation Accuracy in Mechanical Sensor-less Rotor Position Detecting Method of Permanent Magnet Motor using Current Vector Locus” (1995 National Convention of The Industry Applications Society of The Institute of Electrical Engineers of Japan, 180). At step 2, any of the methods can be used and the magnetic pole position is detected with high accuracy. At step 3, a direct current is passed and an IPM motor primary side resistance value is estimated. After that, by using the resistance value obtained at step 3, the motor is started under a sensorless control at step 4. Steps 2 and 3 are carried out in a time as short as possible.

[0058] Since the primary resistance value is used in a control equation in the sensorless control, control characteristics change a little by the primary resistance value. The primary resistance value changes according to the temperature of the IPM motor. Consequently, in order to always obtain the same characteristics, it is necessary to reset the primary resistance value which changes according to the temperature to a proper value. Especially, the starting characteristic is apt to be influenced by the resistance value. By detecting the primary resistance of the IPM motor in short time before starting, an excellent starting characteristic can be always obtained at the time of start in the sensorless control at step 4. That is, a phenomenon such that the motor cannot be started due to an overcurrent irrespective of a temperature change is eliminated.

[0059] A concrete example of the primary resistance estimating method will be explained with reference to FIG. 4.

[0060] In FIG. 4, legend 22 denotes a phase instructing unit which instructs a phase obtained by the detection of the initial magnetic pole position at the time of stop shown in FIG. 3. Legend 23 denotes a Vγ and Vδ instructing unit for instructing that, for example, constant time Vγ=3V (fixed value) and Vδ=0 to the voltage coordinate transformer 16 and, legend 24 denotes a resistance estimating unit A.

[0061] In this case, the resistance estimating unit A 24 estimates the primary resistance R using the following equation (15).

[0062] R=V/γ{square root over ( )}(Iγ²+Iδ²)  (15)

[0063] When the magnetic pole position can be estimated without an error, Iδ=0. Consequently, the following equation (16) may be used in place of the equation (15).

[0064] R=Vγ/Iγ  (16)

[0065] It is sufficient to set estimation time to about five times as long as Ld/R and the estimation time is estimated by the equation (15) or (16) at that time point. By using values obtained via a filter as Iγ and Iδ in the equation (15) or (16), variations in the estimation values can be suppressed.

[0066] Another concrete example of the primary resistance estimating method will be explained with reference to FIG. 5.

[0067] In FIG. 5, legend 25 indicates a γ-axis current controller which performs, for example, a proportion/integration control. An output of the γ-axis current controller 25 is Vγ which is supplied to the voltage coordinate transformer 16. Legend 26 denotes a Vδ instructing unit and gives an instruction that Vδ=0. Legend 27 denotes a resistance estimating unit B. In this case, the resistance estimating unit B 27 estimates the primary resistance R by the following equation (17).

[0068] R=∫(Vγ−R·Iγ)dt  (17)

[0069] The estimation time is set to a value about five times as long as Ld/R. The equation (17) is calculated during the estimation time. With such a configuration, the primary resistance estimation value R is automatically derived so that the integration in the equation (17) becomes zero. That is, by a calculation expression similar to the equation (16), resistance is obtained. With such a configuration, estimation can be excellently carried out without exerting an influence on the estimation value by noise.

[0070] By performing the starting under the sensorless control by using the primary resistance value estimated as described above, an excellent acceleration characteristic can be always obtained irrespective of the temperature of the IPM motor.

[0071] Industrial Applicability

[0072] The apparatus and method for controlling a synchronous motor according to the invention can attain a stable control system by simple computation and is particularly suitable for a synchronous motor which drives without detection of rotation speed and the magnetic pole position.