RU2397600C1 - Drive system of motor with varying magnetic flow - Google Patents

Abstract

FIELD: electricity. ^ SUBSTANCE: drive system of motor with varying magnetic flow includes motor (4) on constant magnets, which has constant magnet, inverter (1) for excitation of motor on constant magnets, and magnetisation device for magnetisation current passage in order to control magnetic flow of constant magnet. Constant magnet is varying magnet the magnetic induction of which is varying depending on magnetisation current from inverter (1). Magnetisation current which exceeds saturation zone of magnetisation of magnetic material of varying magnet is passed through magnetisation device. This system improves repeatability of magnetic flow of varying magnet (53) and accuracy of torque moment. ^ EFFECT: improving control accuracy of torque moment, suppression of transient moment caused by magnetisation process, increasing efficiency and enlarging rotation frequency range. ^ 4 cl, 57 dwg

Description

FIELD OF TECHNOLOGY

The present invention relates to a variable flux motor drive system.

BACKGROUND

Instead of traditional asynchronous motors (IM motors), permanent magnet synchronous motors (PM motors) are distributed, which are highly efficient and are assumed to be miniature and low noise. For example, PM motors have become used as drive motors for electric trains and electric vehicles.

The IM motor creates a magnetic flux through the excitation current from the stator, and therefore has a technical drawback associated with the loss of the passage of the excitation current.

On the other hand, the PM motor is a motor having a rotor equipped with a permanent magnet, whose magnetic flux is used to generate torque, and therefore does not have such a drawback of the IM motor. However, the PM motor induces a speed-dependent voltage with a permanent magnet. In the field of application of electric trains and electric vehicles, which have a wide range of rotational speeds, a condition must be provided under which the induced voltage at the maximum rotational speed does not disable (by electric overvoltage) the inverter, which drives and controls the PM motor. To satisfy this condition, it is necessary to provide an inverter with a sufficiently high withstand voltage or to limit the magnetic flux of the permanent magnet of the motor. The former has an effect on the power source, and therefore the latter is usually chosen. The magnitude of the magnetic flux of the PM motor relative to the magnitude of the magnetic flux of the IM motor (in the IM motor, this is the magnitude of the magnetic flux of the gap created by the drive current) sometimes becomes approximately 1: 3. To create the same torque, a PM motor with a small magnetic flux must pass a large current (torque). Namely, in the low-frequency zone, the PM motor must pass more current than the IM motor in order to produce the same torque.

Accordingly, the maximum permissible inverter current for driving the PM motor must be greater than for driving the IM motor. The switching frequency of the switching element in the inverter of the PM motor is high, and therefore the PM motor causes large losses and heating at a low speed, since the losses are dependent on the current value.

For example, an electric train assumes that it must be cooled by the air flow generated during movement, and if large losses occur during movement at low speed, the inverter should be increased in size to improve cooling capacity. On the other hand, if the induced voltage is high, field attenuation control must be performed. In this case, the superimposed field current worsens the efficiency.

The PM motor has advantages and disadvantages due to the magnet included in it. As an engine, it has the advantage of reducing losses and size. On the other hand, to control the change in the rotation speed of an electric train or an electric car, the PM motor has operating points at which it shows the worst efficiency compared to a traditional IM motor. As for the inverter, increasing the maximum permissible current and losses lead to an increase in the size of the inverter. The efficiency of the system itself is mainly engine dependent, which is why the use of a PM engine improves the overall efficiency of the system. The increase in size of the inverter, however, is a drawback and is not preferred for the system.

FIG. 57 is a block diagram showing an example of a permanent magnet synchronous motor (PM motor) drive system according to the prior art. The main circuit consists of a direct current (DC) power supply 3, an inverter 1 for inverting direct current energy into alternating current energy, and a permanent magnet synchronous motor 4a, which must be driven by the alternating current energy of inverter 1. The main circuit is equipped with a current detection device 2 for detecting motor currents and a rotation angle sensor 18 for detecting a rotation angle of the rotor of the permanent magnet synchronous motor 4a. The inverter 1 inverts the direct current energy from the direct current power source 3 into alternating current energy, which is supplied to the permanent magnet synchronous motor 4a. The currents supplied to the permanent magnet synchronous motor 4a are detected by the current detection device 2 and supplied to the voltage command processing unit 210.

Next, a control operation of this prior art system will be explained. The input to the system is the torque command Tm *. This torque command Tm * is created by the proper means to cause the permanent magnet synchronous motor 4a to produce the required torque. According to the input torque command Tm *, the current command processing unit 211 generates a current command Id * on the D-axis and a current command Iq * on the Q-axis to determine the current on the D-axis and the current on the Q-axis and feeds them to the processing unit 210 voltage commands. The rotation angle of the rotor of the permanent magnet synchronous motor 4a detected by the rotation angle sensor 18 is sent to the voltage command processing unit 210. According to the input D-axis current command Id * and the Q-axis current command Iq *, the voltage command processing unit 210 calculates and creates voltage commands Vd * and Vq * along the D- and Q-axis to pass current so that the current Id along the D-axis and current Iq along the Q-axis were consistent with the current commands. At this time, the voltage command processing unit 210 performs (proportional-integral, PI) PI control for current deviation and receives voltage commands on the D- and Q-axis. After that, the voltage command processing unit 210 converts the coordinates of the voltage commands Vd * and Vq * along the D- and Q-axis and supplies the PWM (pulse width modulation, PWM) circuit 6 with the three-phase voltage commands Vu *, Vv * and Vw *. Although the voltage command processing unit 210 converts the voltage commands along the D- and Q-axis into three-phase voltage commands, for example, a coordinate conversion unit can be adapted to perform voltage command conversion. According to the input commands Vu *, Vv * and Vw * of the three-phase voltage, the PWM circuit 6 controls the on / off (ON / OFF) switching elements of the inverter 1.

As shown in FIG. 57, the drive system of the PM motor of the prior art should have a load contactor 209 on the alternating current side of inverter 1. The permanent magnet synchronous motor 4a has a permanent magnet, and therefore, induces a voltage (counteracting electromotive voltage) when the inverter 1 is locked, while the engine rotates by inertia. If the induced voltage is higher than the DC voltage from the DC power supply 3, an overvoltage is applied to the inverter 1, and a braking force is applied to the synchronous motor 4a.

If the permanent magnet synchronous motor 4a or the inverter 1 causes a short circuit or earth fault, the induced voltage will constantly pass current, causing problems such as overheating and ignition of the permanent magnet synchronous motor 4a and inverter 1. Accordingly, the aforementioned drive system opens load contactor 209, when the inverter 1 is locked to prevent the inverter 1 from receiving the induced voltage, and the permanent magnet synchronous motor 4a and the inverter 1 from the constant th transmission fault current.

The service life of the load contactor 209 is highly dependent on the number of times the open / close operation. When the open / close frequency of the load contactor 209 is high, it suffers from high failure rates and short service life. To improve system reliability, dual load contactors 209a and 209b, 209c and 209d, and 209e and 209f can be set for the respective phases, as shown in FIG. 57. This is not an ideal solution and greatly increases costs.

Unexamined Japanese Patent Application Publication No. H11-299297 (Patent Document 1) describes a technique for controlling the attenuation of a magnetic flux in a permanent magnet motor by decreasing the magnetic flux attenuation current without impairing the accuracy of the torque, thereby reducing inverter and motor losses, as well as the value of the rated current of the device. This, however, allows the excitation current to degrade the efficiency and generate heat. Accordingly, it should have a cooling device that increases the cost and dimensions of the apparatus.

To solve the problems of a conventional permanent magnet synchronous motor drive system, Japanese Patent Publication No. H5-304752, Japanese Patent Document 4, discloses an electric vehicle AC drive motor that uses a combination of a permanent magnet and a field coil to change magnetic flux.

The electric vehicle AC drive motor described in Patent Document 4 effectively engages the motor and inverter in each of the low productivity operation and the high performance operation. This electric vehicle AC drive motor uses a magnetic flux of a permanent magnet embedded in the magnetic pole of the field and, if necessary, a magnetic flux of the field coil to create a magnetic flux of the field. Depending on the output power of the motor, the source of the magnetic flux of the field is switched between only a permanent magnet and both, a permanent magnet and the field winding. At the same time, the excitation current is supplied through a rotating transformer. This electric vehicle AC drive motor can operate in response to engine power output. For example, in response to low power output, it only works with a permanent magnet to improve the operating efficiency. In addition, it can increase the voltage of the motor in the area of low speed of the motor to reduce current, reduce losses in the copper of the motor winding and create losses in the inverter, and improve the efficiency of the system. This effect is significant for an electric car that often drives at low and medium speeds in order to increase current efficiency and extend the travel distance for recharging.

In addition, this electric vehicle AC drive motor does not demagnetize a permanent magnet, and therefore simplifies control of the inverter, does not cause any abnormal overvoltage and protects the system. A rotating transformer can operate at high frequencies to reduce its size, as well as the size and weight of the system as a whole.

The electric vehicle AC drive motor specified in Patent Document 4, however, must always pass the drive current when the magnetic flux is generated by the drive coil. The passage of the excitation current causes a problem of deterioration of the efficiency, and a permanent magnet built into the magnetic pole of the field causes the problem of inducing voltage.

In this regard, there is a variable-flux motor drive system capable of changing the magnetic flux of a magnet by current from an inverter. This system changes the magnitude of the magnetic flux of a permanent magnet according to the operating conditions, and therefore, it is assumed that it should improve the efficiency to a greater extent than the traditional PM-drive system with permanent magnets. When a magnet is not necessary, the magnitude of the magnetic flux can be reduced to minimize the induced voltage.

For a variable-flux motor drive system for driving a variable-flux motor whose magnetic flux is variable controlled by the magnetizing current from the inverter, it is important when and how the magnetization process is performed to change the magnetic flux. A transient torque will occur, depending on the accuracy of the torque or the magnetization process. The magnetization process must be time-controlled to maximize the efficiency and expand the speed range.

The characteristic BH (characteristic of magnetic induction - magnetization) of an alternating magnet shows a steep dependence on the magnetization current from the inverter, for easy fluctuation of the magnetic flux depending on the method of magnetization. Fluctuation of the magnetic flux affects the repeatability of the magnetic flux and the quality of the drive system.

To magnetize a variable magnet, the inverter must pass a large current. In this case, the stator will become saturated. Compared to demagnetizing a variable magnet, magnetizing a variable magnet requires that more current be passed through. The need for a large current increases the maximum permissible current of each switching element of the inverter and, in addition, each switching element of the inverter must have a higher withstand voltage. Namely, only for the magnetization process, each switching element must have a large throughput, which increases its cost. In addition, a large current generates instantaneous heat, and therefore the heat capacity of the inverter must be increased so that the inverter can withstand short-term heat.

Patent Document 1: Unexamined Publication No. H11-299297 of Japanese Patent Application.

Patent Document 2: Publication No. 6800977 of a US patent.

Patent Document 3: Publication No. 5977679 of the US patent.

Patent Document 4: Unexamined Publication No. H5-304752 of Japanese Patent Application.

SUMMARY OF THE INVENTION

The present invention has been made in view of the disadvantages of the aforementioned prior art. An object of the present invention is to provide a variable-flux motor drive system capable of suppressing a deterioration in the accuracy of a torque caused by a variable magnet, suppressing the transient torque created by the magnetization process, improving the efficiency of the system as a whole, and covering a wide frequency range rotation.

Another objective of the present invention is to provide a variable magnetic flux motor drive system for driving a variable magnetic flux motor, whose magnetic flux value is variable controlled by a magnetizing current from an inverter, capable of improving the repeatability of the magnetic flux of a variable magnet and torque accuracy.

Another other objective of the present invention is to provide a variable magnetic flux motor drive system capable of magnetizing an alternating magnet using a simple circuit without increasing the maximum allowable current and heat capacity of the inverter.

Another other objective of the present invention is to provide a variable-flux motor drive system capable of using, instead of a permanent magnet motor, a variable-flux motor whose magnetic flux is variable-controlled, suppressing, as appropriate, counteracting electromotive voltage caused by the magnetic flux of an alternating magnet, using a simple apparatus, preventing the braking force that is applied at high speed tnoj zone, and reliable protection system.

In order to achieve the objectives, an aspect of the present invention provides a variable flux motor drive system having an inverter for driving a variable flux motor that uses a permanent magnet and a variable magnet. The system contains a torque command generation unit to create a torque command for the variable magnetic flux motor, an alternating magnetic flux control unit to change the magnetic flux according to the magnetizing current from the inverter and magnetize the alternating magnet, a switch to pass predetermined current values in D- and the Q axis based on the torque command from the torque command generation unit or the magnetization current commands on the D and Q axis from the variable magnetic flux control unit m, a magnetization request creation unit, to create, if the predetermined conditions are met, a request to the variable flux control unit for magnetizing the variable magnet, and an unlock control command generation unit to create an unlock control command for controlling the inverter according to the setpoint values based on the torque command currents along the D- and Q-axis or magnetization current commands along the D- and Q-axis from the switch.

This aspect of the present invention can suppress the deterioration in the accuracy of the torque due to the variable magnet, suppress the transient torque due to the magnetization process, improve the efficiency of the system as a whole and cope with a wide range of rotational speeds.

Another aspect of the present invention provides a variable flux motor drive system comprising a permanent magnet motor using a permanent magnet, an inverter to drive the permanent magnet motor, and magnetization means to pass magnetization current to control the magnetic flux of the permanent magnet. A permanent magnet at least partially has the ability of a variable magnet, whose magnetic induction is variable in response to the magnetization current from the inverter. The magnetization means transmits a magnetization current that exceeds the saturation zone of the magnetization of the magnetic material of the variable magnet.

Another other aspect of the present invention provides a variable flux motor drive system comprising a permanent magnet motor using a permanent magnet, an inverter to drive the permanent magnet motor, variable flux control means to pass magnetization current to control the magnetic flux of the permanent magnet , means for detecting the current of the motor with permanent magnets, and means for evaluating the magnetic flux to estimate the magnitude of the magnetic otok voltage and current applied to the motor with permanent magnets and the winding inductance serving as a motor parameter. A permanent magnet at least partially has the ability of a variable magnet, whose magnetic induction is adjustable, according to the magnetization current from the inverter.

This aspect of the present invention controls the magnitude of the magnetic flux of the variable magnet according to the magnetization current from the inverter and drives the variable-flux motor, thereby improving the repeatability of the magnetic flux of the variable magnet and the accuracy of the torque.

Yet another aspect of the present invention provides a variable flux motor drive system having an inverter for driving a variable flux motor that has a permanent magnet and a variable magnet. The system includes a main control unit for controlling the inverter, so that the torque of the variable-flux motor corresponds to the torque command, a magnetizing coil for magnetizing the variable magnet of the variable-flux motor, and a magnetization circuit for supplying magnetization current to the magnetizing winding.

This aspect of the present invention can magnetize an alternating magnet using a simple circuit without increasing the maximum allowable current and heat capacity of the inverter.

Another other aspect of the present invention provides a variable magnetic flux motor drive system comprising a permanent magnet motor having a variable magnet which is at least a low coercive force permanent magnet, an inverter for driving a permanent magnet motor, a magnetizing unit for supplying a magnetizing current to control the magnetic flux of the variable magnet, the demagnetization test unit to check whether or not the variable magnet should be demagnetized, and, according to the results of the test, create a demagnetize signal, and one or more demagnetize blocks to demagnetize the variable magnet according to the demagnetize signal created by the demagnetization check unit.

According to the aforementioned and other aspects of the present invention, demagnetization is performed when the variable-flux motor drive system is to be protected or when the inverter is stopped to suppress a counter electromotive voltage, prevent braking force from being applied, and provide reliable protection for the system.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 is a block diagram showing a variable-flux motor drive system according to a first embodiment of the present invention.

FIG. 2 is a simplified model view showing a variable flux motor according to a first embodiment of the present invention.

FIG. 3 is a schematic view showing a rotor of a variable-flux motor according to a first embodiment of the present invention.

FIG. 4 is a characteristic graph showing BH (Magnetic Induction - Magnetization Characteristics) characteristics of a permanent magnet and a variable magnet in a variable flux motor according to a first embodiment of the present invention.

FIG. 5 is a characteristic graph showing a quantitatively correct ratio only in the second quadrant (B> 0, H <0) of FIG. four.

FIG. 6 is a block diagram showing an internal configuration of a magnetization request creating unit according to a first embodiment of the present invention.

FIG. 7 is a block diagram showing an internal configuration of a variable flux control unit according to a first embodiment of the present invention.

FIG. 8 is a timing chart showing an example of an operation of a variable flux motor drive system according to a first embodiment of the present invention.

FIG. 9 is a timing chart showing yet another example of an operation of a variable flux motor drive system according to a first embodiment of the present invention.

FIG. 10 is a block diagram showing a variable-flux motor drive system according to a second embodiment of the present invention.

FIG. 11 is a block diagram showing an internal configuration of a torque command generating unit according to a second embodiment of the present invention.

FIG. 12 is a timing chart showing an example of an operation of a variable flux motor drive system according to a second embodiment of the present invention.

FIG. 13 is a block diagram showing a variable-flux motor drive system according to a third embodiment of the present invention.

FIG. 14 is a block diagram showing an example of an internal configuration of a magnetization request creating unit according to a third embodiment of the present invention.

FIG. 15 is a characteristic graph showing response waveforms when accelerating a variable-flux motor according to a third embodiment of the present invention.

FIG. 16 is a characteristic graph showing response waveforms when decelerating a variable-flux motor according to a third embodiment of the present invention.

FIG. 17 is a block diagram showing another example of an internal configuration of a magnetization request creating unit according to a third embodiment of the present invention.

FIG. 18 is a block diagram showing yet another example of an internal configuration of a magnetization request creating unit according to a third embodiment of the present invention.

FIG. 19 is a block diagram showing a variable flux motor drive system according to a fourth embodiment of the present invention.

FIG. 20 is a block diagram showing an internal configuration of a variable flux control unit according to a fourth embodiment of the present invention.

FIG. 21 is a timing chart showing a variable-flux motor control according to a fourth embodiment of the present invention.

FIG. 22 is a cross-sectional view showing a variable flux motor used by a variable flux motor drive system according to a fifth embodiment of the present invention.

FIG. 23 is a BH characteristic graph showing two variable magnets adopted by a variable flux motor according to a fifth embodiment of the present invention.

FIG. 24 is a block diagram showing an internal configuration of a variable flux control unit according to a fifth embodiment of the present invention.

FIG. 25 is a table of magnetization current viewed by a variable flux control unit according to a fifth embodiment of the present invention.

FIG. 26 is a block diagram showing a variable flux motor drive system according to a sixth embodiment of the present invention.

FIG. 27 is a block diagram showing an internal configuration of a variable flux control unit according to a sixth embodiment of the present invention.

FIG. 28 is a timing chart showing a variable flux motor control according to a sixth embodiment of the present invention.

FIG. 29 is a block diagram showing a variable flux motor drive system according to a seventh embodiment of the present invention.

FIG. 30 is a schematic view showing an example of a rotor in a variable-flux motor according to a seventh embodiment of the present invention.

FIG. 31 is a schematic view showing another example of a rotor in a variable-flux motor according to a seventh embodiment of the present invention.

FIG. 32 is a circuit diagram showing an example of an internal configuration of a magnetization inverter according to a seventh embodiment of the present invention.

FIG. 33 is a circuit diagram showing another example of an internal configuration of a magnetization inverter according to a seventh embodiment of the present invention.

FIG. 34 is a waveform diagram showing examples of waveforms when a magnetizing coil is used for magnetizing according to a seventh embodiment of the present invention.

FIG. 35 is a waveform diagram showing other examples of waveforms when a magnetizing coil is used for magnetizing according to a seventh embodiment of the present invention.

FIG. 36 is a diagram showing an example of contactless power supply from a magnetization circuit to a magnetizing coil of a rotor according to a seventh embodiment of the present invention.

FIG. 37 is a diagram showing another example of contactless power supply from a magnetization circuit to a magnetizing coil of a rotor according to a seventh embodiment of the present invention.

FIG. 38 is a block diagram showing a variable flux motor drive system according to an eighth embodiment of the present invention.

FIG. 39 is a block diagram showing a variable-flux motor drive system according to a ninth embodiment of the present invention.

FIG. 40 is a block diagram showing an inverter used by a variable flux motor drive system according to a ninth embodiment of the present invention.

FIG. 41 is a block diagram showing an example of an internal configuration of a stop / demagnetize check unit used by a variable flux motor drive system according to a ninth embodiment of the present invention.

FIG. 42 shows timing diagrams of the demagnetization control of a variable-flux motor drive system according to a ninth embodiment of the present invention.

FIG. 43 is an explanatory view showing a comparison of magnetic flux control between an existing drive and a variable flux motor drive according to a ninth embodiment of the present invention.

FIG. 44 is a block diagram showing a variable flux motor drive system according to a tenth embodiment of the present invention.

FIG. 45 is a block diagram showing an example of an internal configuration of a stop / demagnetize check unit used by a variable flux motor drive system according to a tenth embodiment of the present invention.

FIG. 46 shows timing diagrams of the demagnetization control of a variable-flux motor drive system according to a tenth embodiment of the present invention.

FIG. 47 is a block diagram showing an example of an internal configuration of a stop / demagnetize check unit used by a variable flux motor drive system according to a tenth embodiment of the present invention.

FIG. 48 shows timing diagrams of a demagnetization control of a variable-flux motor drive system according to a tenth embodiment of the present invention.

FIG. 49 is a block diagram showing a variable-flux motor drive system according to an eleventh embodiment of the present invention.

FIG. 50 shows circuit diagrams of examples of a demagnetization unit capable of conducting demagnetization by shorting the wires of a variable flux motor according to an eleventh embodiment of the present invention.

FIG. 51 is a block diagram showing a variable flux motor drive system according to a twelfth embodiment of the present invention.

FIG. 52 is a flowchart showing the operation of the stop / demagnetize test unit for a variable flux motor drive system according to a twelfth embodiment of the present invention.

FIG. 53 is a block diagram showing a variable-flux motor drive system according to a thirteenth embodiment of the present invention.

FIG. 54 is a flowchart showing the operation of the stop / demagnetize test unit for a variable flux motor drive system according to a thirteenth embodiment of the present invention.

FIG. 55 is a block diagram showing a variable-flux motor drive system according to a fourteenth embodiment of the present invention.

FIG. 56 is a block diagram showing a variable flux motor drive system according to a fifteenth embodiment of the present invention.

FIG. 57 is a block diagram showing a variable-flux motor drive system according to the prior art.

BEST MODE FOR CARRYING OUT THE INVENTION

Below, with reference to the drawings, embodiments of the present invention will be explained in detail.

(First Embodiment)

FIG. 1 is a block diagram showing a variable-flux motor drive system according to a first embodiment of the present invention. First of all, the main circuit including the variable-flux motor 4 of the drive system will be explained. The inverter 1 inverts the direct current energy from the direct current power source into alternating current energy, which is supplied to the variable magnetic flux motor 4. The currents Iu and Iw supplied to the variable-flux motor 4 are detected by the current detection device 2 and converted by the coordinate converting unit 7 into the D-axis current Id and the Q-axis current Iq, which are input to the voltage command processing unit 10. The voltage command processing unit 10 outputs the D-axis voltage command Vd * and the Q-axis voltage command Vq * to the coordinate conversion unit 5, which converts them into three-phase voltage commands Vu *, Vv * and Vw * to be input to the circuit 6 PWM. PWM circuit 6 controls the on / off switching elements of the inverter 1 according to the unlock control command Gst from the unlock control command generation unit 15. The rotation angle 9 of the variable-flux motor 4 is detected by the rotation angle sensor 18 and is differentiated by the pseudo-differentiator 8 into the inverter frequency ω1, which is input to the voltage command processing unit 10 and the magnetic flux command processing unit 12.

Here, a variable flux motor 4 according to a first embodiment of the present invention will be explained. FIG. 2 is a simple model view showing a variable flux motor 4 according to a first embodiment of the present invention. The stator side of the variable-flux motor 4 is similar to that of the existing permanent magnet synchronous motor, and its rotor side, as magnets, has a permanent magnet (FMG) 54, whose magnetic material has constant magnetic induction, and a variable magnet (VMG) 53 whose magnetic material has variable magnetic induction. The permanent magnet synchronous motor has only the permanent magnet mentioned above. A characteristic feature of a variable-flux motor is that it has a variable magnet. Assume that the magnet is magnetized in the direction of the D axis, the permanent magnet 54 and the variable magnet 53 are located in the direction of the D axis. Ld in FIG. 2 is the D-axis inductance, and Lq is the Q-axis inductance.

Then, the permanent magnet 54 and the variable magnet 53 will be explained. The permanent magnet (permanent magnet) maintains a magnetized state without external current being passed to it. Even for a permanent magnet, it is not always true that its magnetic induction is always unchanged under any conditions. Even a permanent magnet synchronous motor is demagnetized or magnetized when excessive current is received from inverter 1. Namely, the permanent magnet is not a magnet whose magnetic flux value is constant, but is a magnet whose magnetic induction is always constant, under normal conditions.

On the other hand, the variable magnet 53 is a magnet whose magnetic induction varies depending on the current passed from the inverter 1, even under standard rated operating conditions. The variable magnet of this variety can be designed within a certain range, according to the material and structure of the magnetic material.

For example, the permanent magnet of the latest permanent magnet synchronous motor is often a neodymium (NdFeB) magnet having a high residual magnetic induction Br. The residual magnetic induction Br is high, for example, about 1.2 T, and therefore, a large torque can be issued from a small structure. Such a magnet is preferred for EV (electric vehicle), HEV (hybrid vehicle) and electric trains that require a small engine with high output power. A traditional permanent magnet permanent magnet motor does not need to be demagnetized by normal current, and therefore, a neodymium magnet (NdFeB) having a high coercive force Hc (about 1000 kA / m) is the optimal magnetic material. A magnet having a large residual magnetic induction and a coercive force is selected for a permanent magnet synchronous motor.

For variable magnet 53, a magnetic material such as AlNiCo (Alnico, Hc = 60 to 120 kA / m) and FeCrCo (Hc = approximately 60 kA / m) having high residual magnetic induction and low coercive value can be considered for use. force. As for the normal amount of current passed from the inverter 1 to drive a conventional permanent magnet synchronous motor, the magnetic induction (magnetic flux) of a neodymium (NdFeB) magnet is substantially constant, and the magnetic induction (magnetic flux) of a variable magnet such like an AlNiCo magnet, is changing. Actually, neodymium is used in the reversible zone, and therefore its magnetic induction can vary in a very small range. However, it returns to its original value if the inverter current stops. On the other hand, the variable magnet 53 is used up to the irreversible zone, and therefore does not return to its original value and retains the changed magnetic characteristic, even after the inverter current is stopped. In FIG. 2, the magnitude of the magnetic flux AlNiCo serving as the variable magnet 53 changes in the direction of the D axis and is substantially zero in the direction of the Q axis.

FIG. 3 is a diagram showing a rotor 51 of a variable-flux motor 4. The permanent magnet 54, for example of NdFeB niode, having a high coercive force, and the variable magnet 53, for example, of AlNiCo, having a low coercive force, are combined and placed on the iron core 52 of the rotor. The direction of the Q axis of the variable magnet 53 and its magnetized direction are orthogonal to each other, and therefore the variable magnet 53 is not influenced by the current along the Q axis and is magnetized by the current 55 along the D axis. 55 - part of the magnetic pole.

FIG. 4 is a characteristic graph showing the BH (magnetic induction-magnetization characteristics) of the permanent magnet 54 and the variable magnet 53. In FIG. 4, S54 is the BH characteristic of the permanent magnet 54, S53 is the BH characteristic of the variable magnet 53, Br1 is the residual magnetic induction of the variable magnet 53, and Br2 is the residual magnetic induction of the permanent magnet 54. In addition, H1sat is the saturation value of the variable magnet 53, Hc1 is the coercive force of the variable magnet 53, and Hc2 is the coercive force of the permanent magnet 54. In addition, X is the magnetization zone due to the output current of the inverter, and Y is the saturation zone of the magnetization of the variable magnet.

FIG. 5 is a characteristic graph showing a quantitatively correct ratio only in the second quadrant (B> 0, H <0) of FIG. 4. In cases of neodymium NdFeB and alnico AlNiCo, there is no significant difference between the residual magnetic inductions Br1 and Br2. As for the coercive forces Hc1 and Hc2, the Alnico AlNiCo magnet is 1/5 to 1/8 of the neodymium NdFeB magnet, and the magnet of FeCrCo is 1/15 of that.

According to the traditional permanent magnet permanent magnet motor, the magnetization zone X with the output current of the inverter 1 is substantially smaller than that of the coercive force of the neodymium NdFeB magnet and is used in the reversible range of its magnetization characteristic. The variable magnet 53 has a small coercive force, as mentioned above, and therefore is used in an irreversible zone, within the range of the output current of the inverter 1, in which it does not return to the magnetic induction B, which it has before applying current, even if the current is zeroed.

An equivalent simplified model of the dynamic characteristics of the variable-flux motor 4 can be expressed using mathematical expression (1). This model is a model in the rotating coordinate system of the DQ axes, with the D axis oriented in the direction of the magnetic flux of the magnet and the Q axis in the direction orthogonal to the D.

[Mat. one]

Here R1 is the winding resistance, Ld is the inductance along the D axis, Lq is the inductance along the Q axis, ΦFIX is the magnitude of the flux (flux linkage) of the permanent magnet, ΦVAR is the magnitude of the flux (flux linkage) of the variable magnet, and ω1 is the frequency of the inverter.

Then, with reference to FIG. 1, control of a variable-flux motor drive system for a variable-flux motor 4 will be explained. Unit 16 generating a work command of the variable-flux motor drive system receives a Run command * of operation and a protection signal PROT from the protection checking unit 17. Based on the Run command Run * and the protection signal PROT from the protection check section 17, the work command creation section 16 creates and displays a Run state flag.

Basically, the Run state flag Run is set to the run state (Run = 1) when the Run * Run command (Run * = 1) is entered. If the run command requires a stop (Run * = 0), the Run flag of the run state is set to the stop state (Run = 0). These conditions occur when the security check unit 17 determines that they are not protective, but normal (PROT = 0). If the protection check unit 17 determines that they are protective (PROT = 1), the operation state is set to the stop state (Run = 0), even if the operation Run * command has the value “1”.

The operation command creating unit 16 generates a torque enable flag TrqON to enable torque application (torque application is enabled if TrqON = 1, and not allowed if TrqON = 0). If the Run Run * command changes from “0” to “1”, the magnetization process is performed. While the magnetization process is in progress, the FCCmp flag to complete the magnetization is set to “1”. Once the magnetization is complete, that is, when the FCCmp magnetization complete flag changes from “1” to “0”, the torque enable flag TrqON becomes “1”. This should increase the torque after magnetization is complete.

Conversely, if the Run Run * command changes from “1” to “0” to give a stop command, the torque enable flag is set as TrqON = 0 to zero the torque command Tm *. After that, the Run state flag Run is set to the stop state (Run = 0). This dependence will be explained later.

If the torque enable flag, TrqON = 0, the torque command generation unit 38 generates a torque command Tm * having a predetermined zero value, and if TrqON = 1, a torque command Tm * having a predetermined torque value. For a transient state of a torque command, the rate of change of the torque command may be limited, or it may undergo a first order system response, depending on the application. In some cases, it can change step by step.

The unlock control command generation unit 15 receives the run state Run flag and generates and outputs the unlock control command Gst to switch elements contained in the inverter 1. If the run state run flag changes from stop (= 0) to run (= 1), create block 15 the unlock control command immediately sets the state (Gst = 1) of the start of the unlock control, and if the Run flag of the operation state changes from start (= 1) to stop (= 0), sets the off state (Gst = 0) after a predetermined time. Turning off the unlock control after a predetermined time should magnetize the variable magnet 53 to a predetermined value, so that the magnet can work freely, as will be explained later.

The magnetic flux command processing unit 12 receives the run state flag Run and the inverter frequency ω1, that is, the rotor speed (ωR), and generates and outputs the flux command Φ *, for example, according to the expression below. This embodiment carries out control by the rotation angle sensor 18 (circular position sensor). Accordingly, the rotation angle sensor 18 detects an angle θ, which is differentiated by the pseudo-differentiator 8 into a rotor speed serving as the output frequency of the inverter (ω1).

[Mat. 2]

Here, Фmin is the minimum value of the magnetic flux (> 0), which must be taken by a variable-flux motor, Фmax is the maximum value of the magnetic flux, which must be taken by a variable-flux motor, and ωA is a predetermined speed.

If the Run state flag Run indicates a shutdown state (Run = 0), the flux command Ф * is set to a minimum Фmin. If it shows the state of operation (Run = 1) and if the rotation frequency ωR is lower than a predetermined value, the flux command Φ is set to a maximum Φmax. If the rotation speed is higher than the predetermined value, the magnetic flux command is set to a minimum Фmin. The setting of the quantities Фmin and Фmах of the magnetic flux will be explained later.

The current setpoint value processing unit 11 receives the torque command Tm * and the magnetic flux command Ф * and calculates the current setpoint IdR along the D-axis and the current setpoint IqR of the current along the Q-axis, as follows:

[Mat. 3]

Here, expressions (3) and (4) are working expressions (the number of pairs of motor poles is one), provided that the reactive torque of the motor 4 with variable magnetic flux is not used. The motor may be any of a pole motor having a difference ΔL between the inductance Ld along the D axis and the inductance Lq along the Q axis, and an implicit pole motor having no such difference

In order to optimize the efficiency or maximize the output power at a predetermined current, the effect of the reactive torque is taken into account. In this case, the following expression is used:

[Mat. four]

Here K is the ratio of the current along the D-axis to the current along the Q-axis, which can vary depending on the aforementioned optimization of the efficiency and maximization of the output power. For optimization, an expression becomes a function whose arguments include torque, speed, and the like, commonly used by simple approximation or as a table.

The magnetization request creation unit 29 sets the magnetization request flag FCReq to “1” if the flux command Φ * or the run state flag Run changes, and to “0” in other cases. FIG. 6 shows the configuration of a magnetization request creation unit 29, in which each of 31 and 33 is a storage unit of a previous value, each of 30 and 34 is a change check unit, and 32 is an OR (OR) circuit.

The previous value of the flux command Φ * is held by the storage unit 31 of the previous value. The output signal from block 31 of the previous value is the magnetic flux command Ф * stored at the last moment of time, and is sent to the change checking unit 30 together with the magnetic flux command value Ф * of this moment in time. The change check unit 30 compares two input signals with each other and, if there is a change between the values of Φ * of the magnetic flux command, outputs “1”. If there are no changes, it displays "0". Namely, it outputs “1” only when the flux command Φ * changes. A similar circuit configuration is prepared for the Run flag of the operation state instead of the flux command Φ *. The output signals from the two change test blocks 30 and 34 are provided to the OR circuit 32, which provides a logical OR function of the two input signals as the magnetization request flag FCReq.

The magnetization request flag FCReq, that is, the output signal from the magnetization request creation unit 29 becomes “1” (magnetization request) if the flux command Φ * or the Run state flag Run changes, and “0” (no magnetization request) in other cases. The Run state run flag, for example, changes when the inverter starts to run, or when it stops, or when it is stopped for protection. Although the embodiment employs a magnetic flux command Φ *, changing the magnetization current command Im * (output from the magnetization current table 27) of the variable flux control unit 13 (to be explained later) can be used to create a magnetization request FCReq.

Then, the variable magnetic flux control unit 13 changes the magnetic flux by the magnetizing current from the inverter and magnetizes the variable magnet. FIG. 7 is a block diagram showing an alternating magnetic flux control unit 13. The variable magnetic flux control unit 13 receives a flux command Φ *, that is, an output signal from the magnetic flux instruction processing unit 12, and calculates and outputs the magnetization current command Idm * on the D-axis and the magnetization current command Iqm * on the Q-axis. This creation process will be explained later.

To magnetize a variable magnet, the command Im * of a predetermined magnetization current is obtained according to the characteristic BH of the variable magnet shown in FIG. 4. In particular, the amplitude of the magnetization current command Im * is set to be equal to or greater than Hlsat, that is, within the magnetization saturation zone Y of the variable magnet 53.

To pass the magnetization current up to the saturation zone of the magnetic material, the magnetic flux values Фmin and Фmах set by the magnetic flux command processing unit 12 are calculated by adding a positive or negative maximum value (saturated) magnetic flux (magnetic induction) of the variable magnet 53 to that of the permanent magnet. Namely, with a positive maximum value of the magnetic flux (the absolute value of the negative maximum value of the magnetic flux is equal to the positive maximum value of the magnetic flux) of the variable magnet, which is ФVARmax, and the magnitude of the magnetic flux of the permanent magnet, which is ФFIX, the following calculations are performed:

[Mat. 5]

Фmin = Ф FIХ - Ф VAR max (7) Фmах = Ф FIX + Ф VAR max (8)

The magnetic flux command Ф * is used to search the magnetization current from the table 27 storing the corresponding magnetization currents and obtain the magnetization current command Im * for the magnetic flux command Ф *. Basically, the magnetization direction of the magnet is in the direction of the D axis, and therefore, the magnetization current Im * is set as the magnetization current Idm * for the axis D.

However, a permanent magnet synchronous motor often causes a difference between the inductance Ld along the D axis and the inductance Lq along the Q axis to produce reactive torque. The transmission of current along the D-axis results in the creation of torque. In particular, this is noticeable in the case of a permanent magnet jet engine (PRM engine), which reaches a wide speed range. Magnetization requires a current value equivalent to or greater than the rated current. Accordingly, the magnetization process will immediately cause a large reactive torque. Even an implicit pole motor will cause partial magnetic saturation when the excessively high magnetization current mentioned above is passed. Namely, the passage of the magnetization current for magnetization causes a transient torsional shock, which creates mechanical vibrations that are not desirable for vehicles and electric trains, since they have a bad effect on comfort.

To avoid this, not only the magnetization current along the D axis must be passed, but also the magnetization current along the Q axis, which is not essential for magnetization, but contributes to a reduction in torque. This inhibits the creation of instantaneous reactive torque. Typically, the torque of a permanent magnet synchronous motor is expressed as follows:

[Mat. 6]

Tm = Ф × Iq + ( Ld-Lq ) × Id × Iq (9)

When the magnetization current Idm is passed along the D-axis, the condition so as not to create a torque is as follows:

[Mat. 7]

Iq = Tm / (Ф + ( Ld-Lq ) × Id ) (10)

Then, using the magnetization current command, the following is obtained:

[Mat. 8]

Iq * = Tm / (Ф * + ( Ld-Lq ) × Idm * ) (eleven)

There is a problem here whether the flux command Φ * should be set before the magnetic flux changes, either after the magnetic flux changes or in between. Magnetization of a variable magnet is complex and, at the same time, uncontrollable. It must be configured on a real machine. For this, the magnetic flux command according to expression (11) can be stored in the table according to the measurements of real machines. Alternatively, D-axis magnetization current commands Idm * and Q-axis magnetization current commands Iqm *, which can reduce discontinuous torque, are obtained by examining real machines and used to prepare a function or table.

In FIG. 7, the magnetization current command processing unit 36 receives the magnetization current command Im * (that is, the magnetization current command Idm * on the D-axis) and generates and outputs the magnetization current command on the Q-axis. The switch 26 outputs the Q-axis magnetization current command Iqm * only when the magnetization complete flag, FCCmp = 1, that is, only during the magnetization process, and when the magnetization complete flag, FCCmp = 0, that is, after the magnetization process is completed, prints zero. According to expression (11), or as mentioned above, the Q-axis magnetization current command processing unit 36 may have a table or function that provides a Q-axis current command to reduce the transient torque actually measured on real machines. The switch 23 operates like the switch 26. When the magnetization complete flag, FCCmp = 1, the switch 23 outputs the D-axis magnetization current command Idm *, and when the magnetization complete flag, FCCmp = 0, it outputs zero.

In FIG. 7, the magnetization request flag FCReq for the request to change the magnetic flux becomes the state (= 1) of the switching request at least instantly. In order to reliably change the magnetic flux, the magnetization request flag FCReq is sent to the minimum pulse enable device 28, which outputs the magnetization complete flag FCCmp (= 1 during magnetization and = 0 when magnetization is complete). This flag has the function not to turn off (= 0) as soon as it turns on (= 1). If the magnetization request flag remains on (= 1) for a predetermined time, the magnetization complete flag FCCmp becomes off as soon as the magnetization request flag FCReq becomes off. Thus, the magnetization process continues only for a predetermined period in order to reliably perform magnetization.

The switch 37 of FIG. 1 receives the FCCmp magnetization complete flag. If this occurs during magnetization (magnetization complete flag, FCCmp = 1), switch 37 outputs the magnetization current commands Idm * and Iqm * along the D- and Q-axis from the variable flux control unit 13 as current commands Id * and Iq * along the D- and Q-axis. If the magnetization is completed (magnetization complete flag, FCCmp = 0), the switch 37 outputs the set current values IdR and IqR along the D- and Q-axis from the current set value processing unit 11 as current commands Id * and Iq * along the D- and Q -axis.

According to the current commands Id * and Iq * along the D- and Q-axis, thus provided, the voltage command processing unit 10 generates voltage commands Vd * and Vq * along the D- and Q-axis to pass currents corresponding to the commands Id * and Iq * current along the D- and Q-axis. The coordinate transforming unit 5 converts the voltage commands into three-phase voltage commands Vu *, Vv * and Vw *, which are used by the PWM circuit 6 to create unlock control signals and the like.

FIG. 8 is a timing chart showing the operation of a variable-flux motor drive system according to the first embodiment. FIG. 8 shows a state in which the protection signal PROT is not set. At time t1, the Run * operation command is entered. Then, the Run state flag Run appears, and the flux command Ф * changes from Ф0 to Фmах. In response to changes in the Run flag Run of the operation state and the magnetic flux command Ф *, the magnetization request flag FCReq appears, and the magnetization complete flag FCCmp remains on for a predetermined time (t1 to t2). Only during the period FCCmp = 1, indicating magnetization, does the D-axis magnetization current command Idm * matter.

When the magnetization completion flag FCCmp changes from “1” to “0” at time t2, the torque enable flag TrqON becomes “1”, and a torque command Tm * is generated from time t2. Namely, the torque command Tm * does not occur until the magnetization completion flag FCCmp becomes 0, upon completion of magnetization, immediately after the start of operation. In this case, a torque command Tm * occurs between time t2 and time t3.

At time t4, the flux command Ф * changes from Фmах to Фmin. Then, the magnetization request flag FCReq appears, and the magnetization completion flag FCCmp occurs and remains on for a predetermined time (t4 to t5). Only during the period in which the magnetization completion flag FCCmp has a value of “1” to indicate magnetization, the D-axis magnetization current command Idm * and the Q-axis magnetization current command Iqm * have values.

Then, at time t6, the Run command Run * stops, the torque enable flag TrqON changes from “1” to “0”, and the torque command Tm * decreases from time t6 to time t7. When the torque command Tm * becomes zero at time t7, the Run state flag Run changes from “1” to “0”, and the magnetic flux command Ф * changes from Фmin to Ф0. In addition, the magnetization request flag FCReq occurs, and the magnetization complete flag FCCmp occurs and maintains the on state for a predetermined time (t7 to t8). Only during the period in which the magnetization complete flag FCCmp is set to “1” to indicate magnetization, does the D-axis magnetization current command Idm * matter.

According to expression (11), the magnetization current command along the Q axis is zero during the magnetization process A (from t1 to t2) and during the magnetization process C (from t7 to t8) and only matters during the magnetization process B (from t4 to t5).

The variable-flux motor 4 transmits an excessively high current during magnetization, and therefore, to some extent creates a transient torque. This torsional shock can cause mechanical vibrations. Such an excessively large current should be minimized to reduce losses and stress state of the elements. To this end, the first embodiment performs a magnetization process under certain conditions to minimize torsional shock, loss and stress state of the switching elements of the inverter 1.

The first embodiment carries out the magnetization process immediately after starting the inverter 1, so that the magnetized state of the variable magnet 53 can be determined before applying a torque. The magnetized state of the variable-flux motor 4 is unknown, for example, immediately after the power source is turned on, and may change due to some reasons while the motor 4 is in the idle state. If the DC voltage of the inverter 1 decreases and if the voltage induced by the common magnetic flux exceeds the DC voltage, the alternating magnets motor passes current to the inverter 1, even if the inverter 1 is stopped. This can change the magnetized state of motor 4. In order to minimize the voltage induced in the idle state, a magnetic flux value that is not used for normal operation can be set for motor 4 before stopping inverter 1. In this case, the magnetic flux required for normal operation mode, should be set for motor 4 when starting the inverter 1.

Since the magnetized state (magnitude of the magnetic flux) is unknown before starting the inverter 1, the first embodiment determines the magnetized state and starts operation. As a result, the first embodiment does not cause any lack of torque due to the insufficient total magnetic flux, or never sets the actual magnetic flux different from the magnetic flux set by the control unit, thereby ensuring the accuracy of the torque.

In addition, the first embodiment calls the torque command Tm * after the magnetization is completed, immediately after starting the inverter. As explained above, passing the magnetization current along the Q axis along with the magnetizing current along the D axis can reduce the transient torque. However, it is unable to accurately determine the magnitude of the magnetization current along the Q-axis, which can correctly suppress the generation of transient torque. On the other hand, only when the torque command Tm * is zero, that is, only when the magnetization current Iq along the Q axis is equal to zero, does the reactive torque occur, which causes a torsional shock, and therefore a smooth start is achieved. The rise of the torque command immediately after starting the inverter 1 and after the magnetization is complete can smoothly start without a torsional shock.

The first embodiment performs the magnetization process after stopping the operation command, that is, after applying the torque becomes optional, and when the magnetization process is completed, closes the valve of the inverter 1.

During high-speed operation, for example, of an electric train, there is an idle state or an inertial state in which the inverter 1 is stopped in order to reduce losses. At this time, if the alternating magnetic flux is large, a large voltage will be induced. If the induced voltage exceeds the DC voltage of the inverter, that is, the voltage of the current collector, the variable-flux motor 4 passes current to the inverter 1, even if the inverter 1 is stopped. This is important for the absorption of kinetic energy from the variable-flux motor 4 and applies braking force to the vehicle. Therefore, this is not preferred. If the magnitude of the speed and magnetic flux is high, the induced voltage may exceed the withstand voltage of the inverter 1 or the sides of the inverter power supply can damage the components. For safety's sake, it is preferable to reset the total magnetic flux and idle.

Thus, a decrease in magnetic flux just before the inverter 1 stops has the effect of suppressing the generation of induced voltage in the idle mode, preventing the application of braking force and avoiding damage to components during overvoltage.

In particular, the first embodiment performs a magnetization process by selecting the magnitude of the magnetic flux, whose absolute value is closer to zero, from among the magnitudes of the magnetic flux expressed by variable magnetic flux and constant magnetic flux. The first embodiment controls three levels of magnetic flux and selects one having the smallest absolute value from among them. This technology can minimize the generation of induced voltage. The same applies to two levels of magnetic flux using one variable magnet 53 and one permanent magnet 54, and to four or more levels of magnetic flux using many variable magnets 53.

The variable magnet 53 may not be used up to the saturation zone. Namely, the variable magnet 53 can be used with an arbitrary intermediate value between a positive maximum and a negative maximum. This, if possible, can reduce the total magnetic flux to zero.

The motor can be designed so that the total magnetic flux of the variable magnet 53 becomes equal to that of the permanent magnet 54. In this case, the variable magnet 53 is used in the saturation zone Y and the total magnetic flux can be reset.

The first embodiment performs magnetization during startup of the inverter and during shutdown of the inverter with a torque command set to zero. This is because the magnetization process passes an excessively large current, and therefore inevitably creates a transient torque. The self-polar jet engine constantly passes current along the Q-axis, and therefore creates a large transient torque due to an excessively large magnetizing current, thereby creating mechanical vibrations and disturbing comfort in the case of an electric train or HEV. In a state of zero torque, even a jet engine does not pass current along the Q-axis, and therefore the creation of a transient torque will be minimized when an excessively large magnetization current is passed in the direction of the D axis.

In the state of the applied torque, also the transient torque is reducible depending on the magnetization current. Initially, the magnetization process is carried out by passing current in the direction of the D axis. In a pole-polar machine, the current along the D-axis changes the torque if there is current on the Q-axis. Even an implicit pole machine often passes an excessively high current, above the rated current, when magnetizing. This can cause instantaneous magnetic saturation, creating a transient torque similar to an explicit pole machine. To suppress this, the current along the Q axis is passed as the magnetization current along with the current along the D axis. The current along the Q axis can suppress transient torque, as can be seen from expression (11).

Thus, the transmission of current along the Q-axis along with the current along the D-axis results in minimizing the transient torque that accompanies the magnetization process, thereby preventing mechanical vibrations, and providing comfort with respect to electric trains and HEV.

The magnetization process suddenly changes the magnetic flux, and an excessively large magnetization current changes the equivalent parameters of the motor. Accordingly, it is difficult to calculate the current along the Q axis in advance, according to expression (11). Therefore, it is necessary to determine the current along the Q-axis on a real machine. Namely, the current along the D axis necessary for the magnetization process is determined, and then the current along the Q axis, such as to suppress transient torque, is determined. Certain currents are stored as data in a table that is used in practice.

FIG. 9 is a timing chart showing another example of the operation of a variable-flux motor drive system according to the first embodiment. In FIG. 9, a protection signal PROT is generated at time ta. Instead of the Run * operation command indicating a stop, the PROT protection signal stops the inverter.

The protection signal PROT detects a protective operation at time ta and changes from “0” to “1”. In this state, the Run Run * command, the Run state run flag, and the Torq enable flag TrqON are simultaneously turned off. The unlock control command Gst leaves the on state to perform the magnetization process. Namely, from ta to tb, the magnetization process is carried out and a magnetic flux command Φ * is provided to bring the absolute value of the magnetic flux as close to zero as possible.

Thus, the inverter 1 does not stop immediately when the protection signal PROT changes from “0” to “1”. Instead, the inverter 1 is locked after a decrease in the total magnetic flux. In a state in which the cause of the protection is unknown, this technology can reduce the likelihood of a secondary failure due to the voltage induced during the closing mode. No protection is necessary to cut off after the magnetization process.

(Second Embodiment)

FIG. 10 is a block diagram showing a variable-flux motor drive system according to a second embodiment of the present invention. The second embodiment is different from the first embodiment shown in FIG. 1, in that the magnetization request creation unit 29 creates the pre-magnetization request flag FCReq0, which must be supplied to the torque command generation unit 38 instead of the magnetization request flag FCReq, which must be supplied to the variable magnetic flux control unit 13, and that the block 38 the creation of the torque commands, in addition to the torque enable flag TrqON, receives the pre-magnetization request flag FCReq0 and the magnetization completion flag FCCmp and, in addition to the torque command Tm * a, creates the magnetization request flag FCReq. Namely, the torque command generation unit 38 prepares a torque command Tm * to reduce the torque during magnetization of the variable magnet, and after the torque reduction, provides the variable magnetic flux control unit 13 with the magnetization request flag FCReq. Accordingly, the variable magnetic flux control unit 13 operates to reduce the torque and then magnetize the variable magnet. The same elements as in FIG. 1 are presented with former references in order not to make repeated explanations.

According to a first embodiment shown in FIG. 1, the output signal from the magnetization request creation unit 29 is the magnetization request flag FCReq. According to a second embodiment shown in FIG. 10, the magnetization request creation unit 29 outputs the pre-magnetization request flag FCReq0. This pre-magnetization request flag FCReq0 is supplied, together with the torque enable signal TrqON and the magnetization complete flag FCCmp, to the torque command generation unit 38.

FIG. 11 is a block diagram showing a torque command generating unit 38 according to a second embodiment. If the flag FCReq0 request pre-magnetization is set to "1", if there is a request to change the magnetic flux, RS-flip-flop 44 is set. The output from the RS flip-flop 44 is sent through the OR circuit 46 to the switch 39. The OR circuit 46 also receives the magnetization complete flag FCCmp.

When the output signal from the OR circuit 46 is “1”, the switch 39 changes its output signal from the set torque value TmR to the predetermined set magnetization torque value TmMag. The switch 40 performs a switching operation in response to the torque enable flag TrqON. If the torque enable flag TrqON has a value of “1”, the switch 40 selects an output signal from the switch 39, and if it has a value of “0”, selects “0”.

If the output from OR circuit 46 is set to “1” and if the switch 39 selects a predetermined magnetization torque set value TmMag with a torque enable flag TrqON in the state of enable (= 1) of the torque, the switch 40 selects and outputs the set value TmMag magnetization torque. The slope of this magnetization torque setpoint TmMag is limited by the change rate limiter 41 to gradually change the torque command Tm *. The comparator 42 checks to see if the torque command Tm * and the magnetization torque command TmMag are equal to each other. If they are equal to each other, “1” is provided, and if not, “0” is provided. The output signal from the comparator 42 and the output signal from the RS flip-flop 44 are supplied to the AND circuit 43, which outputs the logical function AND of the input signals as the magnetization request flag FCReq.

Namely, the rate limiter 41 gradually changes the torque command Tm *. When the torque command Tm * is equal to the magnetization torque command TmMag, the magnetization request flag FCReq is set for the magnetization request (= 1). The magnetization request flag FCReq is also input to a delay circuit 45, which provides an output signal delayed by one control period. The output signal from the delay circuit 45 is supplied to the zero-setting input of the RS flip-flop 44. Namely, the magnetization request flag FCReq arises (= 1) for one control period and then is reset (= 0).

The magnetization request flag FCReq from the torque command generating unit 38 is inputted to the variable magnetic flux control unit 13 to carry out a magnetization process similar to the first embodiment. The magnetization complete flag FCCmp from the variable flux control unit 13 is supplied to the OR circuit 46 of the torque command generation unit 38. As long as the variable magnetic flux control unit 13 is in the magnetization state (magnetization complete flag, FCCmp = 1), the torque command Tm * is set to the magnetization torque command TmMag. As soon as the magnetization is complete (FCCmp = 0), the torque command Tm * gradually increases towards the set torque value TmR under the action of the speed limiter 41.

FIG. 12 is a timing chart showing an example of an operation of a variable flux motor drive system according to a second embodiment. FIG. 12 shows a case in which the magnetization torque command TmMag has a value of 0.

For example, the magnetization process B will be explained (from t4 to t5). According to the flag flag FCReq0 of the pre-magnetization request, the torque command Tm * decreases to zero between time t4 'and time t4. When the torque command Tm * becomes zero at time t4, the magnetization request flag FCReq arises to start the magnetization process. Until the magnetization process is completed (until the FCCmp changes from 1 to 0), the torque command Tm * is supported by the magnetization torque command = 0.

Thus, the magnetization process is performed by reducing the torque command Tm * to the magnetization torque command TmMag. Accordingly, even a jet engine can minimize transient torque when an excessively large magnetizing current is passed in the direction of the D axis, since the current along the Q-axis has a value of zero in the state of zero torque.

(Third Embodiment)

FIG. 13 is a block diagram showing a variable-flux motor drive system according to a third embodiment of the present invention. The third embodiment differs from the first embodiment shown in FIG. 1, in that it additionally has a modulation factor processing unit 61 and a field attenuation control unit 62, and that the output from the field attenuation control unit 62 is supplied to the magnetization request creating unit 29. As a result, the magnetization request creation unit 29 creates a magnetization request based on the modulation coefficient of the inverter 1. The same elements as in FIG. 1 are represented by former references in order not to make repeated explanations.

In FIG. 13, the DC voltage Vdc of the inverter 1 is detected by the DC voltage detection device 60 and supplied to the modulation factor processing unit 61. The modulation factor processing unit 61 receives the DC voltage Vdc and voltage commands Vd * and Vq * along the D- and Q-axis from the voltage command processing unit 10. Based on the voltage commands Vd * and Vq * along the D- and Q-axis, the modulation factor processing unit 61 calculates the amplitude VI * of the voltage command in the coordinate system of the DQ axes, as follows:

[Mat. 9]

Based on the direct current voltage Vdc, the modulation factor processing unit 61 calculates the maximum voltage V1max of the DQ axes, which is provided by the inverter 1, according to the following expression. This maximum output is achieved in single-pulse mode.

[Mat. 10]

The modulation coefficient AL is expressed as mentioned below. The modulation coefficient processing unit 61 outputs the modulation coefficient AL. The modulation factor AL is a value between 0 and 1.

[Mat. eleven]

The field attenuation control unit 62 receives the modulation coefficient AL from the modulation coefficient processing unit 61, and if the modulation coefficient AL is higher than the predetermined maximum modulation coefficient ALmax, corrects and adjusts the currents along the D- and Q-axis so as to equalize AL with ALmax. This is a field weakening control. The predefined maximum modulation coefficient is an arbitrary value and never exceeds 1. An example of the field attenuation control calculates the correction value If * for currents along the D- and Q-axis, as follows:

[Mat. 12]

Here s is the Laplace operator, Kp is the transmission coefficient of the proportional controller, and Ki is the transmission coefficient of the integral controller. If * is enclosed in a region above 0. The above expression controls the compensation current command If * to attenuate the field so that the modulation coefficient AL can be consistent with the maximum modulation coefficient ALmax.

The compensation current command If * for attenuation of the field makes the setpoint values IdR and IqR of the current along the D- and Q-axis applied to the unit 11 for processing the setpoint values of the current, thereby performing the compensation. Namely, compensation is performed by subtracting the correction value from the last stages of expressions (5) and (6) of the block 11 for processing the set current values.

In practice, the If * command of the compensation current is not simply subtracted from the set IdR and IqR of the current along the D- and Q-axis. Instead, the proportions of the D- and Q-axis are determined according to the operating conditions, and then a correction is performed.

Thus, the field attenuation is controlled. The field attenuation control is provided by the If * command of the field attenuation compensation current command with a value equal to or greater than 0 (it is 0 if no field attenuation is performed, and a positive value when field attenuation is performed) to compensate for the set current values for D- and Q-axis.

The field attenuation compensation current command If * is also input to the magnetization request generating unit 29. The magnetization request creation unit 29 creates the FCReq (= 1) flag of the magnetization request in the zone in which the inverter modulation coefficient becomes maximum. FIG. 14 is a block diagram showing an example of a magnetization request creating unit 29 according to a third embodiment. Compared to the magnetization request generating unit 29 according to the first embodiment shown in FIG. 6, the third embodiment further has a comparator 63 for comparing the field attenuation compensation current command If * command with a predetermined set value IfMag, and a change check unit 64 to find the change as a result of the comparison. The same elements as in FIG. 6 are represented by former references in order not to make repeated explanations.

The field attenuation compensation current command If * is compared by the comparator 63 to a predetermined set value IfMag. The If * command of the compensation current increases positively in the field weakening zone, and therefore, the set value of IfMag is set to positive. When the If * command of the compensation current increases to exceed the set IfMag value, “1” is output, and if below the set value, “0” is output. The change check unit 64 detects a point in time when the output signal of the comparator 63 changes, and outputs the detected result to the OR circuit 32. The output signal from the OR circuit 32 is the magnetization request flag FCReq for performing the magnetization process. Namely, the magnetization process is performed when the depth of field weakening becomes large in the field weakening zone.

In the magnetization process, a large flux command Φ * is issued on the low frequency side, and a small flux command Φ * is issued on the high frequency side. For this, the magnetic flux command processing unit 12 can be configured according to the above expression (2). Namely, the flux command Φ * varies depending on the condition of the rotational speed.

FIG. 15 is a characteristic graph showing response waveforms when accelerating a variable-flux motor 4 according to a third embodiment. In FIG. 15, the upper limit of the maximum modulation coefficient ALmax is set to 1. If the rotational speed FR of the variable-flux motor 4 increases since time t0, the modulation coefficient AL increases according to an increase in the rotational speed FR of the variable-flux motor 4. When the modulation factor AL reaches “1” at time t1, field attenuation begins. Namely, the If * command of the field attenuation compensation current gradually increases from zero to positive values. When the field attenuation compensation current command If * reaches IfMag1 at time t2, the magnetization process MG-A is performed to reduce the magnetic flux of the magnet from F1 to F2. At this time, the magnetic flux of the magnet, which is increasing, is still in the WK-A zone of field weakening, and the modulation coefficient remains 1.

When the field attenuation compensation current command If * reaches IfMag2 at time t3, the magnetization process MG-B is performed to reduce the magnetic flux of the magnet from F2 to F3. At this time, the decrease in magnetic flux is large, and therefore, the modulation coefficient AL once decreases below “1” of the field attenuation zone WK-B.

Field attenuation is an essential function to provide a wide range of speed control. However, in the case of a permanent magnet synchronous motor, a large current is needed to weaken the field, which will erroneously reduce the power factor or efficiency. To cope with this, the magnetic flux varies according to the amplitude of the If * command of the field attenuation compensation current, indicating the depth of field attenuation. This suppresses excessive field attenuation current and prevents deterioration of power factor and motor efficiency. This provides an effect of reducing loss and miniaturization of the cooling device. Preventing a reduction in power factor enables the use of switching elements of small maximum permissible currents.

At high speeds, to realize higher speeds of rotation, the weakening of the field continues with a reduced magnetic flux of the magnet. This extends the range of operating speeds.

The third embodiment sets an upper limit on the maximum modulation coefficient ALmax to “1”. This is a single pulse mode. Single-pulse mode outputs the maximum voltage that can be provided by the inverter. This mode includes positive and negative pulses, each of which is provided on one period of the output signal, and therefore the number of switching times per mode period is minimal.

Namely, the single-pulse mode can reduce the switching loss of the switching element, in order to increase the output voltage and reduce the dimensions of the system. In the magnetization process B, the modulation coefficient becomes less than “1” after a change in the magnetic flux. This does not raise any problems for the variable-flux motor drive system. The magnetization process A can maintain a single-pulse mode with a variable value of the magnetic flux of the magnet. In this case, there will be no increase in losses due to optional switching, and the load on the cooling device can be reduced.

It is preferable to carry out the magnetization process before setting the If * command to the field attenuation compensation current, which can automatically magnetize the magnet. If the IfMag value set to switch the field attenuation compensation current If * command is large, the current increasing according to the field attenuation control inadvertently magnetizes to change the magnetic flux of the magnet. Even so, a third embodiment using a variable magnet can realize a wide range of rotational speeds. However, when the present invention is applied to HEVs and electric trains, it is preferable to reduce the torsional shock caused by such a change in magnetic flux. A torsional shock occurs because the control of an alternating magnetic flux dramatically changes the magnetic flux of a magnet. To suppress torsional shock, it is preferable to intentionally change the magnetic flux, for example, by carrying out precise additional current control during the magnetization process.

In FIG. 14, the field attenuation compensation current command If * has only one switching point. Instead, it can have two or more switching points. The example shown in FIG. 15, has two switching points (IfMag1 and IfMag2).

FIG. 16 is a characteristic graph showing response waveforms when decelerating the variable-flux motor 4 of the third embodiment. Like the example of FIG. 15, the magnetization process MG-A generates a magnetization request at time t13 when the field attenuation compensation current command If * becomes smaller than the predetermined value IfMag1. The magnetization process MG-B requires a magnetization process outside the field weakening zone, and therefore is not achievable with a magnetization request based on the field weakening compensation current If *. WK-A and WK-B each indicate a field weakening zone.

To cope with this, the magnetization request creation unit 29 is composed like FIG. 17 instead of the one shown in FIG. 14. FIG. 17 is a block diagram showing another example of a magnetization request creating unit 29 according to a third embodiment. Namely, the magnetization request creating unit 29 receives a modulation coefficient AL instead of the value If * of the field attenuation compensation current command. Comparator 63 compares the modulation factor AL and the predetermined set value ALMag with each other. The change in the comparison result is checked by the change check block 64, which creates a magnetization request for performing the magnetization process B.

FIG. 18 is a block diagram showing yet another example of a magnetization request generating unit 29 according to a third embodiment. This example is different from the example shown in FIG. 17 in that it receives the rotational frequency ωR of the rotor of the variable-flux motor 4 instead of the modulation factor AL of FIG. 17. The rotor rotation frequency ωR is obtained by differentiating the rotation angle θ, which is detected by the rotation angle sensor 18.

The magnetization request generating unit 29 receives the rotational frequency ωR of the rotor of the variable-flux motor 4. The comparator 63 compares the rotor speed ωR and the predetermined set value ωRMag with each other. The result of the comparison is checked by the change checking unit 64, which creates a magnetization request. The example of FIG. 18 implements a wide range of rotational speeds by switching the magnetic flux according to the rotor rotational speed ωR, so that the magnetic flux increases at low rotational speeds and decreases at high rotational speeds.

(Fourth Embodiment)

FIG. 19 is a block diagram showing a variable flux motor drive system according to a fourth embodiment of the present invention. In this embodiment, the variable flux motor, which is a permanent magnet synchronous motor, is the same as that of the first embodiment explained with reference to FIG. from 2 to 5.

According to this embodiment, the main circuit consists of a direct current power source 3, an inverter 1 for inverting direct current energy into alternating current energy, and an alternating magnetic flux motor 4 driven by alternating current energy of inverter 1. The main circuit includes a detection device 2 AC current for detecting engine power; and an angle sensor 18 (position sensor) for detecting engine speed.

Next, a control operation of the above system will be explained. The input signals to it are the Run * operation command and the torque command Tm *. The operation command creating unit 16 receives the operation command Run * and the protection signal PROT from the protection verification unit 17 and generates and displays the Run state flag. Basically, the Run state flag Run is set to the run state (Run = 1) when the run command is entered (Run * = 1), and to the stop state (Run = 0) when the run command indicates the stop (Run * = 0) . When protection is detected (PROT = 1), the run state is set to the stop state (Run = 0), even if Run run * command has the value “1”.

The unlock control command generating unit 15 receives the run state Run flag and generates and outputs the unlock control command Gst to the switching elements included in the inverter 1. The unlock control command generating unit 15 immediately sets the start of the unlock control (Gst = 1) if the Run state run flag changes from stop (Run = 0) to work (Run = 1), and sets the unlock control to be disabled (Gst = 0) a predetermined time after the Run flag of the work state changes from work (Run = 1) to stop (Run = 0).

The magnetic flux command processing unit 12 receives the operation state flag Run and the inverter frequency ω1, that is, the rotor rotation frequency ωR, and generates and outputs the flux command Φ *, for example, according to the expression (16) mentioned below. If it is a stop state (Run = 0), the flux command Ф * is minimized to Фmin. If it is a running state (Run = 1) and if the rotation frequency ωR is lower than a predetermined value, the flux command Φ * is maximized to Φmax. If the rotation speed is higher than the predetermined value, the flux command Φ * is minimized to Φmin.

[Mat. 13]

If ( Run = 0), Ф * = Фmin
Otherwise, if (| ω1 | <ω A ) , Ф * = Фmax
Otherwise Ф * = Фmin (16)

Here, Фmin is the minimum value of the magnetic flux (> 0), which must be taken by the variable-flux motor 4, Фmax is the maximum value of the magnetic flux, which must be taken by the variable-flux motor 4, and ωA is the predetermined speed. The setting of the magnitudes Фmin and Фmах of the magnetic flux will be explained later in connection with the variable magnetic flux control unit 13.

The current setpoint value processing unit 11 receives the torque command Tm * and the magnetic flux command Ф * and calculates the current setpoint IdR along the D-axis and the current setpoint IqR along the Q-axis according to the following expressions (17) and (18):

[Mat. fourteen]

IdR = 0 (17) IqR = Tm * / f * (eighteen)

Expressions (17) and (18) are based on the assumption that the reactive torque of the motor is not used, and the number of poles is 0. The motor can be any explicitly pole motor incorporating the difference ΔL between the inductance Ld along the D axis and the inductance Lq along Q-axis, or an implicit pole motor that does not have this difference.

However, it is effective to take into account the reactive torque in order to optimize the efficiency and maximize the output power at a predetermined current. In this case, the following expression may apply:

[Mat. fifteen]

Here K is the ratio between the current along the D-axis and the current along the Q-axis, is a variable depending on the above optimization of efficiency and maximization of output power. As for optimization, it takes the form of a function with torque, speed, and the like as arguments. The expression can be replaced by a simple approximation or can be prepared in the form of a table. The magnetic flux command Ф * in expression (19) is computed if replaced by the estimated magnetic flux value Фh, which will be explained later.

The structural details of the magnetization request generating unit 29 are similar to those of the first embodiment shown in FIG. 6.

FIG. 20 shows structural details of a variable magnetic flux control unit 13. The variable magnetic flux control unit 13 receives the flux command Φ * from the magnetic flux instruction processing unit 12 and outputs the difference ΔIdm * of the magnetization current along the D-axis to correct the set current value IdR along the D-axis. The difference ΔIdm * of the magnetization current is created according to the sequence of operations mentioned below.

In order to magnetize a variable magnet VMG, the command Im * of a predetermined magnetization current must be calculated according to a characteristic BH of the variable magnet similar to the first embodiment shown in FIG. 4. The amplitude of the magnetization current command Im * is set to be equal to or greater than the Hlsat of FIG. 4, that is, within the saturation zone Y of the magnetization of the variable magnet.

In order to pass the magnetization current up to the magnetization saturation zone Y, the magnetic flux command processing unit 12 sets the magnetic flux Фmin and Фmax by adding a permanent magnet section to the positive or negative maximum value (saturation) of the alternating magnet magnetic flux. If the variable magnet VMG had a positive maximum value ФVARmax of the magnetic flux (the absolute value of the negative maximum is assumed to be equal to that of the positive maximum), and the permanent magnet FMG had the value ФFIХ of the magnetic flux, the following would be proved:

[Mat. 16]

Фmin = Ф FIX - Ф VAR max (21) Фmах = Ф FIХ + Ф VAR max (22)

The magnetic flux command Φ * is inputted to block 13, and the magnetization current table 27 storing the corresponding magnetization currents extracts the magnetization current command Im * corresponding to the magnetic flux command Φ *.

Basically, the magnetization direction of the magnet is in the direction of the D axis, and therefore, the magnetization current command Im * is set to provide the current command Id * along the D-axis. According to this embodiment, the D-axis current setpoint IdR from the current setpoint processing unit 11 is adjusted according to the difference ΔIdm * of the magnetization current command D-axis, thereby providing the current Id * command along the D-axis. Accordingly, the subtractor 26A finds the difference ΔIdm * of the magnetization current along the D-axis, as follows: [Mat. 17]

Δ Idm * = Im * - IdR (23)

To change the magnetic flux, it is also possible to directly apply the magnetization current Im * to the current command Id * along the D-axis.

On the other hand, the magnetization request flag FCReq arises (FCReq = 1) at least instantly when the magnetic flux needs to be changed. To reliably change the magnetic flux, the magnetization request flag FCReq is sent to the minimum turn-on pulse device 28, which outputs the magnetization complete flag (= 1 to indicate that the magnetization is continuing, and = 0 to indicate the completion of magnetization). This magnetization complete flag has a non-disconnect function (= 0) for a predetermined time as soon as it is turned on (= 1). If the input signal is on (= 1) for a period exceeding a predetermined time, device 28 turns off the flag, that is, turns off its output signal.

The switch 23 receives the magnetization complete flag. If magnetization is performed (magnetization complete flag = 1), the switch 23 provides an output signal from the subtractor 26A, and if magnetization is completed (magnetization complete flag = 0), it provides zero.

The voltage command processing unit 10 of FIG. 19 receives current commands Id * and Iq * along the D- and Q-axis and generates voltage commands Vd * and Vq * along the D- and Q-axis, so that currents corresponding to the commands can be passed.

The voltage commands Vd * and Vq * along the D- and Q-axis from the voltage command processing unit 10 are converted by the coordinate conversion unit 5 into three-phase voltage commands Vu *, Vv * and Vw *. According to the three-phase voltage commands, the PWM circuit 6 creates an unlock control signal for the PWM control of the inverter 1. The coordinate conversion unit 7 converts the alternating current currents Iu and Iw detected by the current detection device 2 into the detected current values Id and Iq along the D- and Q-axis which are input to the voltage command processing unit 10. The pseudo-differentiator 8 differentiates the angular signal θ from the angle sensor 18 (position sensor) to the frequency ω1 of the inverter. The voltage command processing unit 10, the coordinate transformation units 5 and 7, and the PWM circuit 6 are known.

FIG. 21 is an example of a timing chart showing the effect of various signals. In this example, the protection signal is not set (PROT = 0). According to the change in the Run flag Run of the operation state or the change of the flux command Φ *, the magnetization request flag is set, and the magnetization completion flag is set to guarantee the magnetization request flag for a predetermined time. During the period in which the magnetization complete flag is set, the magnetization current command Im * has a value.

This embodiment with the above configuration provides the results mentioned below. The variable flux motor 4 shows a sharp change in the characteristic relating to the magnetizing current of the inverter, as shown in characteristic BH of FIG. 4. Accordingly, in practice it is difficult to re-receive the same magnetic flux, even if the same control is repeated, since there is an axial deviation between the D axis and the magnetic flux axis, caused by control without a position sensor, the difference in the requested current and the individual difference engine. Poor repeatability of the magnetic flux impairs the accuracy of the torque, and therefore is not desirable.

To this end, the variable-flux motor drive system of the embodiment transmits a higher magnetization current than the magnetization saturation zone of the magnetization characteristic of the VMG variable magnet to magnetize the variable magnetic flux after magnetization, improve repeatability, ensure torque accuracy and improve drive reliability.

The variable-flux motor drive system of the embodiment sets the minimum time for passing the magnetization current, and therefore, magnetization will not be completed completely. This keeps the magnitude of the variable magnetic flux after the magnetization process from changing and improves the accuracy of the torque.

(Fifth Embodiment)

With reference to FIG. 22 to 25, a variable flux motor drive system according to a fifth embodiment of the present invention will be explained. FIG. 22 shows a design of a variable flux motor 4A to be controlled by a variable flux motor drive system according to a fifth embodiment of the present invention. The variable flux motor 4A of this embodiment differs from the variable flux motor 4 of the fourth embodiment in that the variable magnet VMG consists of a pair of two different permanent magnets with a low coercive force.

Namely, the rotor 51 has an iron rotor core 52, which includes combinations of magnets, each consisting of a high coercive force permanent magnet 54, such as an NdFeB neodymium magnet, a low coercive force permanent magnet A 53, and a low permanent magnet B 57 coercive force, and the magnet A 53 and magnet B 57, for example, are alnico (AlNiCo) magnets and form a pair. The high coercive force magnet 54 is an FMG permanent magnet and is mounted in the magnetic pole portion 55 in a direction orthogonal to the diameter of the iron core of the rotor 52. The low coercive force magnet A 53 is a variable magnet VMG and is mounted on the other side of the magnetic pole portion 55 in the diameter direction of the rotor iron core 52. Namely, the magnet A 53 is located in the boundary region between adjacent parts 55 of the magnetic poles. Another permanent magnet B 57 with a low coercive force is installed in parallel with a permanent magnet 54 with a high coercive force on the center side of the rotor 51. The two permanent magnets A 53 and magnet B 57 with a low coercive force are made of the same magnetic material, i.e. alnico ( AlNiCo) similar to the fourth embodiment.

Thus, each variable magnet VMG consists of a permanent magnet A 53 and magnet B 57 with a low coercive force, which are made of the same magnetic material and installed in different positions. Accordingly, two constants, magnet A 53 and magnet B 57, with a low coercive force, provide different magnetization actions relative to the magnetization current along the D axis. This forms the structure of a variable magnet having two BH characteristics shown in FIG. 23.

FIG. 23 shows different characteristics of BH, that is, curves C53 and C57 of variable magnets. Two curves C53 and C57 of variable magnets are provided by alnico magnets mounted in different spatial positions. The use of a pair of two permanent magnets with a low coercive force from different materials can also provide two BH characteristics. Magnets of the same magnetic material have the same BH characteristic. However, depending on their position in the engine, they exhibit different values of Ф of the magnetic flux with respect to magnetization H by current. The characteristics shown in FIG. 23 show the relationship between the magnetization current from the inverter and magnetic flux and are based not only on materials. LV-A - level A of the inverter magnetization current setting, and LV-B - level B of the inverter magnetization current setting.

According to this embodiment, the design of the variable flux motor 4A is different from that of the fourth embodiment, as mentioned above, and the setting of the magnetizing current amplitude for changing the magnetic flux is different from that of the first embodiment. According to this embodiment, the configuration of the variable-flux motor drive system is similar to that of the fourth embodiment shown in FIG. 19. The functional configuration of the variable magnetic flux control unit 13 is shown in FIG. 24 and differs from that of the fourth embodiment. Details of the variable magnetic flux control unit 13 according to this embodiment will be explained.

A variable magnet with a small coercive force is referred to as a variable magnet A, and a variable magnet with a high coercive force is referred to as a variable magnet B. Two magnetization current commands Im_A and Im_B are provided step by step.

Im_A is in the saturation zone of the magnetization, that is, above the HcAsat of the variable magnet A and in the reversible zone of the variable magnet B.

Im_B is in the saturation zone of the magnetization, i.e. above HcBsat, with regard to both variable magnets B and A.

The variable magnetic flux control unit 13 selects one of two magnetization current commands (a positive-negative combination is possible) according to the level of the command Ф * of the requested magnetic flux and provides the magnetization current command Im *.

The variable magnetic flux control unit 13 calculates a magnetization current command Im * according to the magnetic flux command Φ *. The variable magnetic flux control unit 13 of this embodiment is different from that of the fourth embodiment shown in FIG. 20 in that a block 35 for storing the previous value is added, and the magnetization current table 27 has two arguments. The microcomputer control is used to repeat the control process at predetermined intervals.

The flux command Φ * and the magnetization request flag FCReq are input to the previous value storage unit 35. Block 35 stores the flux command Φ * at each rising edge of the magnetization request flag. The output from the previous value storage unit 35 is the value of the flux command Φ * provided when the magnetization request flag FCReq has increased to 1 last time, that is, the value of the flux command Φ * for the previous magnetization process. Here, the previous value of the magnetic flux command is indicated by reference as Φ * old. The magnetization current table 27 receives the value Φ * of the magnetic flux command of this point in time and the previous value Φ * old of the magnetic flux command.

The magnetization current table 27 is set as shown in FIG. 25. The maximum magnetic flux of the variable magnet A 53 is VARAmax, and the maximum magnetic flux of the variable magnet A 57 is VARBmax. Then, the flux command can take the following four values:

[Mat. eighteen]

( i ) F1 = Ф FIX - Ф VARА max - Ф VARВ max
( ii ) Ф2 = Ф FIХ + Ф VARA max-Ф VARB max ( iii ) Ф3 = Ф FIX -Ф VARA max + Ф VAR max
( iv ) Ф4 = Ф FIX + Ф VAR max + Ф VARB max (24)

As is apparent from the table of FIG. 25, the magnetization current table 27 is characterized in that even the same magnetic flux can adhere to a different magnetization process, that is, a different magnetization current depending on the previous state. For example, in the table of FIG. 25, the case with the previous value of the magnetic flux command, Ф * old = Ф2, and the current value of the magnetic flux command, Ф * = Ф3, corresponds to "Im_B → -Im_A". This means that the magnetization process is first performed with Im * = Im_B, and then continues with Im * = -Im_A. This can be achieved simply by changing the magnetization current command Φ * from Im_B to -Im_A according to time. In order to reliably carry out the magnetization process, it is performed with Im_B, and after the magnetization is complete, similar to the fourth embodiment or the sixth embodiment, which will be explained later, it continues by changing the flux command Ф * to -Im_A and again setting the magnetization request flag.

According to this embodiment, the variable-flux motor drive system uses two or more variable magnets A 53 and magnet B 57 of different characteristics to set four levels of magnetic flux values instead of two that are achievable with a single variable magnet. In particular, this embodiment sets the magnetization current within the reversible magnetization zone or the saturation zone, and therefore each alternating magnet has a specific value. Accordingly, this embodiment can establish a magnetic flux that is reproducible and has good repeatability in order to improve the accuracy of the torque. The ability to apply many levels of magnetic flux allows the magnetic flux to be finely tuned depending on operating conditions. Thus, the embodiment makes the best use of a variable flux motor to improve the efficiency of the system. Although an embodiment employs a combination of two variable magnets A and B, a combination of three or more variable magnets is also possible.

(Sixth Embodiment)

FIG. 26 shows a variable flux motor drive system according to a sixth embodiment of the present invention. In FIG. 26, parts common to the fourth embodiment shown in FIG. 19 are represented by the same reference.

The variable flux motor drive system according to the sixth embodiment is different from the fourth embodiment shown in FIG. 19 in that the magnetic flux estimating unit 9 is additionally installed for evaluating the magnetic flux Фh using voltage commands Vd * and Vq * from the voltage command processing unit 10, currents Id and Iq along the D- and Q-axis from the coordinate conversion unit 7 and angular rotor rotation frequency ω1 and outputs the estimated magnetic flux to the variable magnetic flux control unit 13. In addition, the variable flux control unit 13 has the configuration shown in FIG. 27.

The magnetic flux estimating unit 9 uses the commands Vd * and Vq * of the voltage along the D- and Q-axis, the currents Id and Iq along the D- and Q-axis and the angular frequency ω 1 of the rotor rotation (inverter frequency) to estimate the magnitude of the magnetic flux along D -axis as follows:

[Mat. 19]

Ф h = (Vq * -R 1 × Iq - ω1 × Ld × Id - Lq × dIq / dt) / ω 1 (25)

The estimated magnetic flux value Фh and the magnetic flux command Ф * from the magnetic flux command processing unit 12 are input to the variable magnetic flux control unit 13.

A detailed configuration of the variable magnetic flux control unit 13 of this embodiment is shown in FIG. 27. The subtractor 19 calculates the deviation between the magnetic flux command Φ * and the estimated magnetic flux Φh and generates a deviation in the PI (proportionally integral) PI controller 20. The magnetic flux command Φ * is input to the magnetizing current setpoint processing unit 21. The magnetization current setpoint value processing unit 21 calculates the magnetization current command Im * using a table or a functional equation so that magnetization is performed to realize the magnetic flux corresponding to the magnetic flux command Φ *. At this time, a BH characteristic similar to that mentioned above is used. The adder 22 adds the output signal from the block 21 for processing the set value of the magnetization current to the output signal from the block 20 PI control.

The adder 22 provides a magnetization current command Im *. For magnetization, the magnetization current command Im * is used as the current Id * command on the D-axis. To force Id * and Im * to be consistent with each other, an embodiment uses a subtractor 26A to subtract the D-axis current setpoint IdR from the magnetization current command Im * and provides a D-axis magnetization current command difference ΔIdm *. This difference is added to the given value IdR of the current along the D-axis by the adder, and therefore the command Id * of the current along the D-axis is consistent with the command Im * of the magnetization current.

The switch 23 in the variable magnetic flux control unit 13 selects one of two input signals according to the magnetization completion flag, which will be explained later, and outputs the magnetization current command Idm *. If the magnetization completion flag has the value 0 (magnetization completed), the difference ΔIdm * of the magnetization current command on the D-axis is set to zero. If the magnetization complete flag has a value of 1 (during magnetization), the output signal from the adder 22 is provided as ΔIdm *.

The deviation between the magnetic flux command Φ * from the subtractor 19 and the estimated magnetic flux Φh is input to the magnetization completion check unit 24. The magnetization completion check unit 24 provides 1 if the absolute value of the magnetic flux deviation is less than the predetermined value α, and if it is larger than α, it provides 0. Trigger 25 (RS-FF) receives the magnetization request flag FCReq at input S installation unit, and the output signal from the block 24 checks to complete the magnetization at the input R of the installation of zero. The output from RS-FF 25 is the magnetization complete flag, which is input to the PI control unit 20 and switch 23. If the magnetization complete flag is 0, it indicates that magnetization has been completed, and if 1, it indicates that magnetization is being performed .

The estimated value hh of the magnetic flux from the magnetic flux estimating unit 9 is also input to the current set value processing unit 11. The current setpoint value processing unit 11 uses the estimated magnetic flux value Фh instead of the magnetic flux command Ф * used by the expression in the first embodiment, and calculates the current setpoint values IdR and IqR along the D- and Q-axis, as follows:

[Mat. twenty]

With the above configuration, this embodiment provides the actions and results mentioned below. If there is a magnetization request, the magnetization request flag becomes 1, at least immediately. RS-FF 25 is then set to provide a magnetization complete flag of 1 to indicate that magnetization is ongoing. The switch 23 provides, as a command Im *, the magnetization current, the output signals from the PI controller 20 and the processing unit 21 of the set value of the magnetization current. The magnetization current setpoint processing unit 21 proactively provides the magnetization current based on a predetermined characteristic BH so as to achieve a flux command Φ *. This results in the instantaneous execution of magnetization up to the approximate value of the command. Since the magnetization time is reduced, the generation of excessive torque or loss can be suppressed. The BH characteristic can be obtained in advance through testing.

As mentioned above, it is difficult to make the magnetic flux exactly match the set value. Accordingly, this embodiment corrects the magnetization current Im *, as shown in FIG. 28 so that the deviation | Ф * - Фh | magnetic flux was approaching 0 due to the action of the PI controller 20 in the variable magnetic flux control unit 13. At the last stage, this causes the flux command Φ * and the estimated magnetic flux Φh (that is, the actual magnetic flux, if there is no estimation error) to be consistent with each other. This improves the repeatability of the magnetic flux during the magnetization process and increases the accuracy of the torque.

According to this embodiment, the magnetization completion check unit 24 outputs 1 to the variable flux control unit 13 if the absolute value of the magnetic flux deviation falls within a predetermined range α, as shown in FIG. 28 to indicate that a predetermined magnetic flux value has actually been achieved, and magnetization has been completed. In response to a zeroing request, RS-FF 25 provides the magnetization completion flag FCCmp at 0. Thus, when the estimated magnetic flux value is consistent with the magnetic flux command Φ *, the magnetization process is completed. As a result, this embodiment improves the repeatability of the magnetic flux during the magnetization process and increases the accuracy of the torque.

According to this embodiment, the target values IdR and IqR of the current along the D- and Q-axis are created using the estimated value Фh of the magnetic flux estimated from voltages and currents, and therefore, the current commands along the D- and Q-axis are corrected according to actual conditions, even if the magnitude of the magnetic flux changes due to the magnetization process. Since currents on the D- and Q-axis are passed according to the current commands on the D- and Q-axis, the effect of changes in the magnitude of the magnetic flux on the torque can be reduced and the accuracy of the torque can be improved.

This embodiment is based on the estimated value of the magnetic flux. The estimated magnetic flux value takes into account motor inductances such as Ld and Lq. These values vary according to magnetic saturation. In particular, a variable magnetic flux motor changes its magnetic saturation depending on the magnitude of the variable magnetic flux. Therefore, to improve the accuracy of estimating the magnetic flux and the accuracy of the torque, it is useful to use a function or table that takes the estimated value of the variable magnetic flux and displays the inductance of the motor.

Even if such a table is prepared, it is sometimes difficult to correctly capture the inductance characteristics. In this case, the magnetic flux estimate can be replaced by a magnetic flux detection device consisting of Hall elements and the like, for detecting the real magnetic flux Fr and using it instead of the estimated magnetic flux value Phh. This will further improve the accuracy of the magnetic flux estimate and the accuracy of the torque.

(Seventh Embodiment)

FIG. 29 is a block diagram showing a variable flux motor drive system according to a seventh embodiment of the present invention. First of all, the main circuit including the variable-flux motor 4 of the drive system will be explained. The inverter 1 inverts the direct current energy from the direct current power source 3 into alternating current energy, which is supplied to the variable magnetic flux motor 4. The currents Iu and Iw supplied to the variable-flux motor 4 are detected by the current detection device 2 and input to the coordinate conversion unit 7 of the main control unit 122. The coordinate transformation unit 7 converts them into a current Id along the D-axis and a current Iq along the Q-axis, which are input to the voltage command processing unit 10.

The voltage command processing unit 10 issues a D-axis voltage command Vd * and a Q-axis voltage command Vq * to a coordinate conversion unit 5, which converts them into three-phase voltage commands Vu *, Vv * and Vw *, which are input to circuit 6 PWM PWM circuit 6 controls the on / off switching elements of the inverter 1 according to the unlock control command Gst from the unlock control command generation unit 15. The rotation angle of the variable-flux motor 4 is detected by the rotation angle sensor 18 and is differentiated by a pseudo-differentiator 8, which provides an inverter frequency ω1 to be input to the voltage command processing unit 10 and the magnetic flux command processing unit 12.

The magnetic flux command processing unit 12 receives the operation command Run and the inverter frequency ω1, i.e., the rotor rotation frequency ωR (the angle detected by the rotation angle sensor 18 and differentiated by the pseudo-differentiator 8 is the rotor speed, i.e., the inverter output frequency), and calculates the command Ф * magnetic flux. The current setpoint value processing unit 11 receives the torque command Tm * and the magnetic flux command Ф * and calculates the current setpoint IdR along the D-axis and the current setpoint IqR of the current along the Q-axis. The variable magnetic flux control unit 13 changes the magnetic flux according to the magnetization current from the inverter 1, thereby magnetizing the variable magnet. For this, block 13 creates a magnetization correction amount ΔIdm * based on the flux command Φ *. The magnetic flux command Ф * is added by the adder 14 to the current D-axis current value IdR in order to prepare the current D-axis current command Id *, which is input to the voltage command processing unit 10. The voltage command processing unit 10 also receives the current Iq along the Q axis as the current command Iq * along the Q axis.

The variable magnetic flux motor 4 has a magnetizing coil, which will be explained later, for receiving the magnetizing current from the magnetization circuit 128. The magnetization circuit 128 consists of a magnetization inverter 120, a direct current power source 127, and a magnetization current control unit 123. The magnetization current control unit 123 has a current control unit 119. The magnetization winding is connected through a contact ring 121 to the inverter 120 magnetization circuit 128 magnetization. The magnetization inverter 120 of the magnetization circuit 128 is connected to a DC power source 127 to supply the magnetization current to the magnetization coil. The current passed to the magnetization coil is detected by the current detection device 124. The current control unit 119 generates a release control signal for the magnetization inverter 120 so that the detected magnetization current can match the magnetization current command value ImagRef.

The variable flux motor 4 of this embodiment is similar to that of the first embodiment and is structured like the simple model shown in FIG. 2. However, the design of the rotor 51 is different from that of the first embodiment shown in FIG. 3, and is as shown in FIG. thirty.

The rotor 51 of this embodiment shown in FIG. 30 includes a combination of a permanent magnet 54, for example of NdFeB niode, having a high coercive force, and a variable magnet 53, for example, of AlNiCo alnico, having a low coercive force, the combination is mounted on the iron core 52 of the rotor. The magnetized direction of the variable magnet 53 is orthogonal to the direction of the Q axis, and therefore free from the influence of current along the Q axis and can be magnetized by current along the D axis. The rotor 51 has a magnetization winding 56, which receives current from the magnetization circuit 9, so that its magnetic field can be directed to act on the variable magnet 53.

The characteristics BH (magnetic induction-magnetization characteristics) of the permanent magnet 54 and the variable magnet 53 are similar to those of the first embodiment shown in FIG. 4 and 5.

Then, with reference to FIG. 29, parts of a characteristic of a variable-flux motor drive system for a variable-flux motor 4 according to this embodiment will be explained. The nodes characteristic of the variable-flux motor drive system are a magnetic flux command processing unit 12, a magnetic flux request generating unit 17 and an alternating magnetic flux control unit 13.

The magnetic flux instruction processing unit 12 calculates a magnetic flux (permanent flux magnetic flux + variable magnet flux) that are suitable for the current situation. The magnetization request creation unit 17 sets the magnetization request flag FCReq to request the magnetization process to magnetize (magnetize or magnetize) an alternating magnet when the flux command Φ * from the magnetic flux instruction processing unit 12 has changed or when the inverter enters an operation state (when Run * changes from 0 to 1).

If the magnetization request flag FCReq is set, the variable magnetic flux control unit 13 performs a magnetization process to bring the magnitude of the magnetic flux after magnetization to the magnetic flux command Φ *. This is achieved by adding the magnetization correction value ΔIdm * to the predetermined current value IdR along the D-axis to provide a current command Id * along the D-axis. The voltage command processing unit 10 calculates the voltage command Vd * along the D-axis so that the current along the D-axis can follow the current command Id * along the D-axis. The so-called PI current regulation or the like is included in the voltage command processing unit 10.

This embodiment is characterized in that the rotor 51 of the variable speed motor 4 is provided with a magnetization coil 56, as shown in FIG. 30, and the magnetization coil 56 receives the magnetization current from the magnetization circuit 128 through the contact ring 121 (FIG. 29).

First of all, the magnetization winding 56 provided for the variable-flux motor 4 will be explained. As shown in FIG. 30, a magnetizing coil 56 is mounted adjacent to the alternating magnet 53 of the rotor 51. A single-phase magnetizing coil 56 is connected via a contact ring 121 to the inverter 120 of the magnetization of the magnetization circuit 128, as shown in FIG. 29. The magnetization inverter 120 is connected to a DC power supply 127 to supply a magnetization current Imag to the magnetization coil 56. The current passed to the magnetization winding 56 is detected by the current detection device 124 and inputted to the current control unit 119. The current control unit 119 generates a release control signal for the magnetization inverter 120 so that the detected magnetization current can match the magnetization current command value ImagRef.

The current passed to the magnetization winding 56 does not need regeneration and requires that the magnetization current Imag be passed in the positive and negative directions (for magnetization and demagnetization), and therefore, the magnetization inverter 120 is implemented by the known single-phase two-half-wave bridge inverter. The current control unit 119 controls the current through PI control so that the magnetization current Imag can be matched with the magnetization current command value ImagRef. With regard to the passage of magnetization current Imag, it must be done quickly and accurately. Accordingly, PI control can be replaced by a hysteretic comparator. The magnetization current command ImagRef is generated by the variable magnetic flux control unit 13, similar to the prior art.

This configuration can pass a predetermined magnetization current Imag to the magnetization coil 56 integrated in the rotor 51 to directly control the magnetization of the alternating magnet 53 by the magnetic field generated by the magnetization current Imag. Compared with the prior art, which passes an excessively large current along the D-axis to the main magnetizing coil, the magnetizing current Imag to the magnetizing coil 56 can alternately control the magnetic flux of the variable magnet 53.

This results in a decrease in the maximum permissible current of the inverter 1, thereby reducing the size, weight and cost of the inverter 1. The magnetic flux of the magnetization coil 56 built into the iron core 52 of the rotor is synchronized with the magnetic flux of the magnet, and therefore there is no magnetic flux that is coupled to magnetization coil 56 and varies in time. As a result, the magnetization winding 56 does not create a counter electromotive voltage, and therefore, the DC voltage of the magnetization inverter 120 can be significantly lower than that of the inverter 1. This allows the adoption of small-sized high-current elements to reduce the permissible performance of the magnetization inverter 120.

If the inverter 1 is used to magnetize the variable magnet 53, the variable magnet 53 and the permanent magnet 54 reinforce each other. In this case, the magnetic flux of the stator side is saturated, so that the magnetic field created by the magnetizing current, if it is passed, has almost no effect on the variable magnet. Compared to the demagnetization of the variable magnet 53, the magnetization of the variable magnet 53 needs to pass an excessively large magnetization current into the stator winding. This results in an increase in the nominal performance of each element of the inverter 1, which increases the cost and dimensions of the system.

To cope with this, an embodiment sets up a magnetization coil 56 on the rotor 51 to eliminate the effect of magnetic saturation. With a minimum current from the magnetization inverter 120, the alternating magnet 53 can be easily magnetized.

If an excessively large magnetizing current is passed as the stator current along the D-axis, its transient response to the current can produce a transient torque. In particular, if the variable-flux motor is clearly polar, passing the current along the D-axis can change the reactive torque to change the torque of the variable-flux motor 4. Namely, passing an excessively large magnetization current as the current along the D-axis causes an excessively large change in torque. Such a change in torque can create mechanical vibrations in the case of a vehicle and increase the risk of a bad effect on the system. In the case of a vehicle, passengers may feel uncomfortable.

To cope with this, the embodiment only passes the required minimum magnetizing current to minimize the generation of transient torque. In particular, when the motor is clearly polar, the current passed to the magnetization winding 56 will not directly cause the reactive torque. Compared with the prior art, which uses the inverter 1 for magnetization from the stator side, an embodiment can appreciably dampen torque changes due to magnetization.

When passing the magnetization current from the inverter 1, a voltage margin is required. Typically, the output voltage varies according to speed. For high-frequency rotation, a single-pulse mode is adopted to increase the output voltage and reduce losses. When the inverter 1 operates at the maximum output voltage, there is no margin for passing a positive magnetization current, that is, a positive current along the D-axis, in order to increase the magnetic flux of the alternating magnet 53. If there is no voltage margin, the magnetization current slowly increases to stretch the period for passing current. During this period, the temperature of the inverter 1 rises, causing a risk of system damage. To avoid this, the cooler must be reinforced.

To cope with this, this embodiment does not create any counter electromotive voltage on the magnetization winding 56 and does not cause a state in which no margin of magnetization voltage is present due to other factors similar to the case of inverter 1. An embodiment may optionally apply voltage at any time to pass magnetization current. If inverter 1 is used for magnetization, there will be no degree of freedom of inductance. The inductance of the motor is designed taking into account the output power and efficiency of the motor, and therefore the optimization of magnetization is not always given high priority. On the other hand, the use of an exclusive use magnetization coil 56 increases the degree of freedom of the designed inductance to realize the optimum inductance for magnetization.

In the above explanation, the magnetizing coil 56 generates a magnetic flux to magnetize the variable magnet 53. To this end, the magnetizing coil 56 is mounted near the variable magnet 53, as shown in FIG. 30. Instead, as shown in FIG. 31, the magnetizing coil 56 can be installed near the permanent magnet 54 so that the magnetic flux generated by the magnetizing coil 56 can neutralize the magnetic flux of the permanent magnet 54.

In the rotor 51 shown in FIG. 31, a magnetization coil 56 is mounted adjacent to the permanent magnet 54, which is a magnet with high coercive force. This setup attenuates the magnetic flux of the permanent magnet 54 by passing a magnetizing current into the magnetizing coil 56.

As explained above, the problem caused when the inverter 1 passes the magnetizing current is that the stator side is magnetically saturated when the alternating magnet 53 is magnetized. In order to completely magnetize the alternating magnet 53, an excessively high stator current must be passed. Passing such an excessively large current can cause torque fluctuations during magnetization. In addition, increasing the maximum permissible current of the inverter 1 raises the problem of increasing the cost and volume of the system.

To cope with this, the embodiment of FIG. 31 sets a magnetization coil 56 to attenuate the magnetic flux of the permanent magnet 54. The magnetization coil 56 acts on the permanent magnet 54 to neutralize the magnetic flux of the permanent magnet 54. This reduces the magnetic saturation of the stator side, so that the inverter 1 can pass less magnetization current to magnetize the alternating magnet 53.

In this case, the magnetization inverter 120 can be simplified to operate in only one quadrant to pass only the negative magnetization current. For example, as shown in FIG. 32, the magnetization inverter 120 may consist of one switching element Tr and one diode D. On the other hand, if the magnetization current is to be passed in both directions, a half-wave inverter capable of operating in two quadrants is required. A half-wave inverter, for example, must have four Tr switching elements, as shown in FIG. 33 to allow magnetization inverter 120 to pass positive and negative currents to magnetization coil 56. Thus, an embodiment can simplify the magnetization inverter 120 to reduce the size and cost of the system.

FIG. 34 is a waveform graph showing examples of waveforms when the magnetization coil 56 is used for magnetization. The case shown in FIG. 34, performs magnetization not only with the magnetization coil 56, but also with the inverter 1. When the magnetic flux command Φ * calculated by the magnetic flux instruction processing unit 12 changes, the magnetization request creating unit 17 instantly sets the magnetization request flag FCReq to “ one". Taking the magnetization request flag FCReq, the variable magnetic flux control unit 13 performs the magnetization process in a predetermined time Ta. The magnetization process provides the magnetization current command value ImagRef for the magnetization coil 56 and the D-axis current command Id *, which is the current command for magnetization of the main control unit 122 of the inverter 1.

The magnetization current Imag and current Id on the D-axis of the inverter 1 are used to control and transmit currents. Magnetization current Imag and D-axis current Id create magnetic fields to change a variable magnet. In FIG. 34, the magnetization between times t1 and t2 of time magnetizes the variable magnet 53, and the magnetization between times t3 and t4 of time demagnetizes the variable magnet 53.

Compared to the demagnetization side, the magnetization side is not light due to magnetic saturation. Accordingly, the magnetization is carried out simultaneously from the magnetization winding 56 and the inverter 1. On the other hand, the demagnetization side is lighter than the magnetization side, and therefore only the magnetization winding 56 is used.

Magnetization through the inverter 1 more or less causes a fluctuation in torque, and therefore is not preferred. If the magnetization coil 56 alone is ineffective for magnetization, the inverter 1 can be used in conjunction with the magnetization coil 56 to easily carry out magnetization.

FIG. 35 is a waveform graph showing another example of a magnetization process. The magnetization coil 56 acts only on the magnetization side. As mentioned above, the demagnetization side is lighter than the magnetization side, and therefore, is successfully performed by the inverter 1 without involving the magnetization circuit 128, that is, without passing the magnetization current to the magnetization coil 56. In this case, the magnetization inverter 120 can be simplified to operate in a single quadrant, as shown in FIG. 32. This contributes to reducing the size and weight of the system. The magnetization inverter valve 120 is turned on only during the magnetization process (from t1 to t2 and from t3 to t4).

The supply of energy from the magnetization circuit 128 to the magnetization coil 56 of the rotor 51 can be achieved in a non-contact manner. FIG. 36 is a diagram showing such a case. In FIG. 29, the magnetization inverter 120 supplies energy through the contact ring 121 to the magnetization coil 56. In FIG. 36, the rotary transformer 132 supplies energy to the magnetization winding in a non-contact manner. The rotary transformer 132 has a power supply side winding 130 mounted on the stator and a power receiving side winding 129 mounted on the rotor 51.

The magnetization inverter 120 applies a high frequency voltage corresponding to the magnetization current command to the magnetizing coil 56. The power receiving coil 129 of the rotary transformer 132 induces the same voltage. The power receiving side coil 129 has a neutral point and a rectification circuit 133 for conducting half-wave rectification. The output of the rectification circuit 133 is connected to a magnetization winding 56, which receives a diode-converted DC current and transmits a magnetization current to change the magnetic flux of the alternating magnet 53.

Compared to slip ring 121, this embodiment can magnetize in a non-contact manner to improve maintenance performance and reliability. The circuit on the receiving side of the energy includes only diodes that are mounted on the rotor 51, which can increase the temperature. However, advanced semiconductor technology, such as SiC (silicon carbide), may not cause temperature problems. As mentioned above, the magnetization winding 56 needs little voltage, and a rotating transformer does not need a large space.

In this example, current is passed to the magnetization coil 56 in one direction. When the inverter 1 is used for magnetization, it is not necessary to pass an excessively high stator current, since the stator will not be magnetically saturated, as shown in the waveforms of FIG. 34. If the magnetizing current is passed in both directions, as shown in FIG. 35, the DC output side of rectifier 133 may have a second magnetization inverter 134, as shown in FIG. 37. This configuration can pass the magnetization current in both directions through the magnetization winding 56. The magnetization circuit 128 can be performed on the rotor 51 of the variable-flux motor 4. In this case, the slip ring 121 and the rotary transformer 132 are optional.

(Eighth Embodiment)

With reference to FIG. 38, a variable flux motor drive system according to an eighth embodiment of the present invention will be explained. According to a seventh embodiment shown in FIG. 29, one inverter 1 drives one variable-flux motor 4. On the other hand, the variable-flux motor drive system of this embodiment drives four variable-flux motors 4 through one inverter 1. FIG. 38, parts common to those of the seventh and other embodiments are represented by common references.

The main control unit 122 of the inverter 1 is composed similar to that of FIG. 29. Each variable-flux motor 4 is provided with a magnetization circuit 128 including a magnetization current control unit 123 and a magnetization inverter 120. Typically, a DC power supply 127 is installed. The magnetization current control unit 123 further has a stabilizing regulator 126, whose output signal is added to the magnetization current command ImagRef. The stabilizing controller 126 receives the number of revolutions WR (x) of the variable-flux motor 4 (x), differentiates it, multiplies the result by the transfer coefficient, and outputs the product to correct the magnetization current command ImagRef.

Typically, the excitation of multiple synchronous motors by one inverter 1 can cause vibration and destabilize the number of revolutions of synchronous motors if the load on the motors 4 with variable magnetic flux is unbalanced. Inverter 1 does not have the means to stabilize individual synchronous motors, and therefore can lose synchronism and become inoperative. Because of this, there are several examples for the simultaneous excitation of synchronous motors. A system using one inverter for one synchronous motor, that is, one variable-flux motor 4, is higher in cost than a drive system capable of simultaneously driving multiple, for example, asynchronous motors.

On the other hand, the embodiment shown in FIG. 38, has a stabilizing regulator 126 that can pass magnetization current to suppress a change in the speed of the variable-flux motor 4 caused by uneven load distribution. This stabilizes the speed of each variable-flux motor 4.

The torque of the variable-flux motor 4 is determined by the magnetic flux of the magnet and the stator current. The stator current is general and unregulated. By increasing the magnetic flux of the alternating magnet 53 using the magnetizing coil current 56, the total magnetic flux becomes adjustable. To stabilize the speed, the speed deviation can return to the torque (since it is equivalent to internal friction). This is realized by creating feedback on the derivative of the number of revolutions. Providing a magnetization coil 56 for each variable flux motor 4 may realize the simultaneous operation of the variable flux motors 4, which are synchronous motors. Such simultaneous operation is unattainable by the prior art. An embodiment centrally places the inverter 1, to reduce the size and cost of the system.

As explained above, the magnetization winding 56 provided for the rotor 51 does not create a counter electromotive voltage, and therefore the required voltage is small. Basically, applying a voltage corresponding to the resistance of the winding is sufficient. Compared to magnetizing with inverter 1, an embodiment can very easily magnetize an alternating magnet 53 and can always individually control torque using magnetization coil 56.

(Ninth Embodiment)

FIG. 39 is a block diagram showing a configuration of a variable flux motor drive system according to a ninth embodiment of the present invention. In FIG. 39 and in the embodiments described below, elements that are the same as or equivalent to those of the embodiments already explained are represented by the same reference so as not to make repeated explanations. The permanent magnet synchronous motor, i.e. the variable flux motor 4, is the same as that of the first embodiment shown in FIG. from 2 to 5.

The variable flux motor drive system according to this embodiment includes a variable flux motor 4, a current detection device 2, a direct current power supply 3, an inverter 1 for converting direct current energy to alternating current energy, a switch 205a, a PWM circuit 6 , a stop / demagnetize test unit 208a, a voltage command processing unit 210, a current command processing unit 211, and a rotation angle sensor 18. The variable-flux motor drive system is divided into a main circuit and a control circuit. A DC power source 3, an inverter 1, a variable magnetic flux motor 4, a current detection device 2 for detecting a motor current, and a rotation angle sensor 18 for detecting a rotation angle of the variable magnetic flux motor 4 form a main circuit. A PWM selector switch 205a 6, a stop / demagnetize test unit 208a, a voltage command processing unit 210 and a current command processing unit 211 form a control circuit. The variable-flux motor 4 has a variable magnet (e.g. an alnico magnet), which is a permanent magnet with a low coercive force.

The inverter 1 drives the variable-flux motor 4. In addition, the inverter 1 corresponds to the magnetization unit of the present invention in order to supply a magnetization current for controlling the magnetic flux of the variable magnet of the variable-flux motor 4. In addition, the inverter 1 corresponds to the demagnetization unit of the present invention for demagnetizing the variable magnet according to the demagnetization signal generated by the stop / demagnetize checking unit 208a. The inverter 1 is connected directly to the variable-flux motor 4 and, unlike the prior art, does not need a load closure. According to an embodiment, there is one demagnetization unit. You can install many demagnetization blocks. An embodiment using a plurality of demagnetization blocks will be explained later.

The stop / demagnetize check unit 208a corresponds to the demagnetize check unit of the present invention, which checks whether the variable magnet of the variable-flux motor 4 is demagnetized and, according to the result of the test, generates a demagnetize signal. If the inverter 1 stops operating or if a failure occurs inside or outside the variable-flux motor drive system, the stop / demagnetize check unit 208a determines that the alternating magnet should be demagnetized and generates a demagnetize signal to stop the system for protection.

The voltage command processing unit 210, the current command processing unit 211, and the rotation angle sensor 18 are the same as those of the prior art shown in FIG. 57, and therefore, repeated explanations are omitted.

The switch 205a switches the output signal to another according to the demagnetize signal generated by the stop / demagnetize test unit 208a. If no demagnetize signal is output from the stop / demagnetize test unit 208a (demagnetize flag, FLG_DEMAG = 0), the switch 205a provides a 6-phase PWM circuit with three-phase voltage commands Vu *, Vv * and Vw * generated by the voltage command processing unit 210.

On the other hand, if the stop / demagnetize check unit 208a outputs a demagnetize signal (demagnetize flag, FLG_DEMAG = 1), the switch 205a outputs 0. In this case, the PWM circuit 6 provides the inverter 1 with a control signal so that phases U, V and W at the same time, ON / OFF operations were repeated. The inverter 1, therefore, shorts the wires of the variable-flux motor 4 to demagnetize the variable magnet.

FIG. 40 shows the detailed configuration of inverter 1. As mentioned above, the stop / demagnetize test unit 208a outputs a demagnetize signal to simultaneously turn on or off the switching elements of the three phases. Therefore, the inverter 1 shorts the wires of the variable-flux motor 4 to demagnetize the variable magnet 53. One way to demagnetize the variable magnet 53 of the variable-flux motor 4 is to turn on one of the six switching elements of inverter 1. Turning on one of the switching elements can lead the rotor 51 to a predetermined phase angle, so that the demagnetization current can be passed to demagnetize the induced voltage. The induced voltage of the variable-flux motor 4 causes a problem if the variable-flux motor 4 is rotating. Due to rotation, the rotor 51 necessarily crosses a predetermined phase angle of rotation, and therefore the alternating magnet 53 can be demagnetized.

Another way is to reduce the output voltage of the inverter 1 to achieve demagnetization. The shorting of the wires of the motor 4 with variable magnetic flux is equivalent to zeroing the output voltage of the inverter 1. The demagnetization is achieved sufficiently only by reducing the output voltage. For example, like a normal magnetization process, the magnetization current command necessary to achieve the target value of the demagnetization magnetic flux is specified as a current command along the D axis to pass the demagnetization current for demagnetization. The output voltage decreases in response to a decrease in the magnetic flux of the magnet, that is, the demagnetization of the magnet, depending on the period for the passage of the magnetization current.

When the demagnetization unit of the present invention shorts at least one inter-wire connection of the variable-flux demagnetizing motor 4 and when the demagnetization current is at a predetermined level, the shorting period can be very short. An instantaneous short circuit will suffice.

An operation of an embodiment having the above configuration will be explained. FIG. 41 is a block diagram showing a detailed configuration of a stop / demagnetize test unit 208a. The input signals to the block are the PROT protection signal and the operation command RUN_CMD. These signals are created by the proper means in the system. Based on these signals, the stop / demagnetize test unit 208a receives a time reference when the inverter 1 stops operation, or a time reference when the variable-flux motor drive system needs to be protected.

Basically, when an operation command is entered, the operation command is in the operation state (RUN_CMD = 1), and when the operation command sets a stop, the operation command is in the stopped state (RUN_CMD = 0).

First of all, a normal stop will be explained. FIG. 42 (a) shows the states of signals along the time axis during a normal stop. In normal operation, the operation command is RUN_CMD = 1, and the protection signal is PROT = 0. Accordingly, the circuit 220 DOES NOT output 1, and the circuit 221 And also outputs 1. Here, the output signal from the circuit 221 And is considered as an operation command taking into account protection.

If normal operation continues, the storage circuit 223 of the previous value outputs 1. Since the circuit 222 DOES NOT output 0, the circuit 224 AND outputs 0. By reception 1, the OFF time delay circuit 225 (OFFTD) outputs 0 after a predetermined time. In this example, 0 is fed continuously to the OFFTD circuit 225, and therefore, the 225 OFFTD circuit continuously outputs 0. As a result, the demagnetization flag FLG_DEMAG = 0. OR circuit 226 outputs 1.

A serious failure test circuit 227 receives, through appropriate means, the state of the variable-flux motor drive system. If the variable-flux motor drive system is stable, or is in a minor failure state, or in a state of other device malfunction, the severe-failure test circuit 227 outputs 0. If the variable-flux motor drive system has a serious failure, the serious-failure test circuit 227 outputs 1. In this example, the system is stable, and therefore, a serious failure test circuit 227 outputs 0. Accordingly, circuit 228 DOES NOT output 1, and circuit 229 AND outputs 1.

As a result, the stop / demagnetize check unit 208a outputs the unlock control command Gst to 1. Based on the unlock control Gst command generated by the stop / demagnetize test unit 208a, the PWM circuit 6 controls (controls unlock) the switching elements included in the inverter 1.

At time t0 of FIG. 42 (a), the operation command sets the stop (RUN_CMD = 0), the OFFTD circuit 225 takes 1, and the demagnetize flag appears (FLG_DEMAG = 1). At time t0, the unlock control command Gst remains 1. Since the demagnetize flag has a value of on, the switch 205a outputs 0, as mentioned above. In this case, the PWM circuit 6 provides the inverter 1 with a control signal so that the phases U, V and W simultaneously repeat the on / off operations. Therefore, the inverter 1 shorts the wires of the variable-flux motor 4 to demagnetize the variable magnet.

After a predetermined time elapses from time t0, the OFFTD circuit 225 outputs 0 at time t1. As a result, the demagnetization flag disappears (FLG_DEMAG = 0). Only during the interval (from t0 to t1) in which the demagnetization flag has a value on, the switch 205a outputs 0, and the inverter 1 performs demagnetization.

At time t1, the unlock control command Gst becomes 0. Accordingly, the switching elements of the inverter 1 are locked, and the inverter 1 stops operation.

Then, a protective stop in the event of a minor failure will be explained. FIG. 42 (b) shows signal states along the time axis during a protective stop in the event of a minor failure. At time t0, the protection signal PROT changes from 0 to 1. Scheme 221 And displays an operation command that takes protection into 0. Then, OFFTD circuit 225 outputs 1, and the demagnetize flag (FLG_DEMAG = 1) appears to perform demagnetization by inverter 1. After passing a predetermined time, the unlock control command Gst becomes 0 time t1. As a result, the switching elements of the inverter 1 are locked, and the inverter 1 stops working. What is different from a normal shutdown is that the run command saves the run state (RUN_CMD = 1), and the protection signal PROT becomes 1 to demagnetize and stop the inverter 1.

In conclusion, a protective stop will be explained in the event of a serious failure. FIG. 42 (c) shows the signal states along the time axis during a protective stop for a serious failure. The criteria for minor failure and serious failure may be optionally set by the designer or user. Usually, a serious failure is a failure that belongs to the system, it is dangerous if the valves are turned on according to the unlock control command, and it is necessary that the system be stopped immediately. At time t0, the protection signal PROT changes from 0 to 1. At the same time, the serious failure test circuit 227 outputs 1 to indicate that it is a serious failure. As a result, the unlock control command Gst becomes 0 to immediately stop the inverter 1.

At time t0, the OFFTD circuit 25 outputs 1, and a demagnetization flag appears (FLG_DEMAG = 1). However, the inverter 1 is stopped, and therefore, no demagnetization is performed.

As explained above, the variable-flux motor drive system of this embodiment performs demagnetization when the variable-flux motor drive system is to be protected or when the inverter 1 is stopped. This suppresses counteracting electromotive voltage, prevents braking power and reliably protects the system.

When the stop / demagnetize test unit 208a outputs a demagnetize signal, the inverter 1 is controlled to simply perform demagnetize. Accordingly, existing devices can be effectively used to implement this system. Performing demagnetization can suppress the opposing electromotive voltage to eliminate the load contactor 209 shown in FIG. 57, thereby reducing costs.

Although this embodiment directly connects the inverter 1 and the variable-flux motor 4 to each other, a contactor for controlling the electrical connection can be installed between the inverter 1 and the variable-flux motor 4, similar to the prior art. In this case, the demagnetization test unit 208a outputs a control signal for opening the contactor when the inverter 1 stops operating, or when the variable-flux motor drive system needs to be protected. This configuration improves the reliability of the variable-flux motor drive system. This variable-flux motor drive system has an inverter 1, serving as a demagnetization unit, and therefore each phase can have only one contactor (instead of a dual arrangement).

FIG. 43 shows a comparison of magnetic flux control between an existing drive and a variable flux motor drive of an embodiment. In FIG. 43, 251a and 251b are the withstand voltages of the inverter, 252a and 252b are the voltages at the motor terminals, 253a and 253b are rotational speeds, 254a and 254b are the magnitude of the magnetic flux of the magnet, and 255a and 255b are counteracting electromotive voltages. Tcst - time reference of inertia movement.

FIG. 43 (a) shows the magnetic flux control of an existing drive, i.e. a permanent magnetic flux permanent magnet (PRM) synchronous motor. The magnetic flux 254a of the magnet at PRM is constant regardless of the rotational speed, and therefore, the generated counter electromotive voltage 255a rises according to the increase of the rotational speed 253a. In the case of an electric train, EV, HEV or ship, many drive systems drive a single object. In this case, each drive is not able to independently determine the speed (engine speed) of the object. An external force (wind, incline, and the like) acting on an object can accelerate the object. In this case, the rotational speed will increase even if the inverter 1 is stopped, so that the motor rotates by inertia, and the counter electromotive voltage 255a rises according to the rotational speed. As explained above, it is likely that the counter electromotive voltage 255a exceeds the withstand voltage 251a of the inverter 1, which will damage the drive, create a braking force applied to the motor and cause a short circuit failure.

To cope with this, the variable-flux motor drive of the embodiment shown in FIG. 43 (b), performs demagnetization when the inverter 1 is stopped, in order to minimize the magnetic flux 254b of the magnet. Even if the rotation frequency 253b increases when the variable-flux motor 4 rotates by inertia, the counteracting electromotive voltage 255b is suppressed in order to reliably protect the system. In the low-frequency zone, the magnetic flux of the magnet 254b may increase to reduce the current passed to the variable-flux motor 4, thereby reducing the size and cost of the inverter 1.

(Tenth Embodiment)

FIG. 44 is a block diagram showing a configuration of a variable flux motor drive system according to a tenth embodiment of the present invention. This embodiment differs from the ninth embodiment in that no switch 205a is installed, contactors 207a and 207b are installed between the wires of the variable-flux motor 4, and the stop / demagnetize test unit 208b outputs a demagnetize signal to control the contactors 207a and 207b.

The switches 207a and 207b correspond to the demagnetization unit of the present invention, and, based on the demagnetization signal generated by the stop / demagnetize test unit 208b, shorten the wires of the variable magnet, thereby performing demagnetization. This embodiment, therefore, includes two demagnetization units. Similar to the first embodiment, the stop / demagnetize check unit 208b determines that the variable magnet 53 should be demagnetized if the inverter 1 stops working or if the variable-flux motor drive system is to be protected and generates a demagnetize signal to be supplied to the contactors 207a and 207b. The remaining configuration of the embodiment is the same as that of the ninth embodiment, and therefore, repeated explanations are omitted.

Then, the operation of the embodiment with the above configuration will be explained. FIG. 45 is a block diagram showing a detailed configuration of a stop / demagnetize test unit 208b. The input signals to it are the PROT protection signal and the operation command RUN_CMD. These signals are created by the proper means in the system.

The work command is basically set to the work state (RUN_CMD = 1), and when the work command sets a stop, the work command is set to the stop state (RUN_CMD = 0).

First of all, a normal stop will be explained. FIG. 46 (a) shows the states of signals along the time axis during a normal stop. In normal operation, the rotation command is RUN_CMD = 1, and the protection signal is PROT = 0. Accordingly, the demagnetize flag, FLG_DEMAG = 0. The stop / demagnetize check unit 208b outputs the unlock control command Gst to 1. The PWM circuit 6 receives the Gst command the unlock control generated by the stop / demagnetize test unit 208b, and accordingly controls the switching elements included in the inverter 1.

At time t0 of FIG. 46 (a), the operation command sets the stop (RUN_CMD = 0), the 225 OFFTD circuit accepts 1, and the demagnetize flag turns on (FLG_DEMAG = 1). At this time, the stop / demagnetize test unit 208b outputs a demagnetize signal to the contactors 207a and 207b. Based on the demagnetization signal, the contactors 207a and 207b short-circuit the wires of the variable magnet, thereby conducting demagnetization.

Unlike the ninth embodiment, the unlock control command Gst becomes 0 at time t0. Accordingly, the switching elements of the inverter 1 are locked, and the inverter 1 stops operation. According to a ninth embodiment, the inverter 1 serves as a demagnetization unit, and therefore, no demagnetization is performed if the inverter 1 is stopped. The variable flux motor drive system of the tenth embodiment employs contactors 207a and 207b as demagnetization units, and therefore can perform demagnetization even if the inverter 1 stops.

After a predetermined time from time t0, OFFTD circuit 225 outputs 0 at time t1. As a result, the demagnetization flag turns off (FLG_DEMAG = 0), and the contactors 207a and 207b stop the demagnetization by the inter-wire shorting circuit.

Then, a protective stop in the event of a minor failure will be explained. FIG. 46 (b) shows signal states along the time axis during a protective stop for a minor failure. At time t0, the protection signal PROT changes from 0 to 1. OFFTD circuit 225 outputs 1, and the demagnetize flag appears (FLG_DEMAG = 1), so that the contactors 207a and 207b perform demagnetization. At the same time, the unlock control command Gst becomes 0. This locks the switching elements of the inverter 1, and the inverter 1 stops the operation. What is different from a normal shutdown is that the run command saves the run state (RUN_CMD = 1), and the PROT protection signal becomes 1 to stop the inverter 1.

In conclusion, a protective stop will be explained in the event of a serious failure. FIG. 46 (c) shows signal states along a time axis during a protective stop for a serious failure. At time t0, the protection signal PROT changes from 0 to 1. At the same time, the serious failure test circuit 227 outputs 1 to indicate that it is a serious failure. As a result, the unlock control command Gst becomes 0, and the inverter 1 stops operation immediately. According to an embodiment using the stop / demagnetize test unit 208b, the variable-flux motor drive system acts similarly for a minor failure and a serious failure.

FIG. 47 is a block diagram showing yet another example of a configuration of the stop / demagnetize check unit 208b. First of all, a normal stop will be explained. FIG. 48 (a) shows the states of signals along the time axis during a normal stop. In normal operation, the operation command is RUN_CMD = 1, and the protection signal is PROT = 0. Accordingly, the demagnetize flag, FLG_DEMAG = 0. The stop / demagnetize check unit 208b outputs the unlock control command Gst to 1. The PWM circuit 6 receives the Gst command the unlock control generated by the stop / demagnetize test unit 208a, and accordingly controls the switching elements included in the inverter 1.

At time t0 of FIG. 48 (a), the operation command sets the stop (RUN_CMD = 0), and the demagnetize flag turns on (FLG_DEMAG = 1). At this time, the stop / demagnetize test unit 208b outputs a demagnetize signal to the contactors 207a and 207b. Based on the demagnetization signal, the contactors 207a and 207b short-circuit the wires of the variable magnet 53, thereby conducting demagnetization. At time t0, the unlock control command Gst becomes 0, and therefore, the switching elements of the inverter 1 are locked to stop the inverter 1. After that, the demagnetize flag remains on (FLG_DEMAG = 1) to continue demagnetization. The unlock control command Gst remains 0.

Then, a protective stop in the event of a minor failure will be explained. FIG. 48 (b) shows signal states along the time axis during a protective stop for a minor failure. At time t0, the protection signal PROT changes from 0 to 1. As a result, the demagnetize flag is turned on (FLG_DEMAG = 1), so that the contactors 207a and 207b perform demagnetization. At the same time, the unlock control command Gst becomes 0. This locks the switching elements of the inverter 1 to stop the inverter 1.

In conclusion, a protective stop will be explained in the event of a serious failure. FIG. 48 (c) shows signal states along the time axis during a protective stop for a serious failure. At time t0, the protection signal PROT changes from 0 to 1. At the same time, the serious failure test circuit 227 outputs 1 to indicate that it is a serious failure. As a result, the unlock control command Gst becomes 0 to immediately stop the inverter 1.

As mentioned above, the variable flow motor drive system according to the tenth embodiment has contactors 207a and 207b serving as demagnetization units, and therefore is capable of performing demagnetization after the inverter 1 stops operation, in addition to the result of the ninth embodiment. The prior art shown in FIG. 57 must have a load contactor connected in series for each phase, and therefore must have at least three (six for dual line-up) load contactors 209a through 209f. On the other hand, in the embodiment, only two contactors are needed. A series-connected load contactor, provided for each phase, must always pass a large current during operation of the inverter, and therefore must have a large permissible load. According to an embodiment, the contactors 207a and 207b only transmit current for a short demagnetization period. Accordingly, according to the embodiment, the contactors may each have a reduced maximum permissible current, a reduced size, and a reduced failure rate.

The use of the stop / demagnetize test unit 208b shown in FIG. 47, reduces the number of internal circuits, dimensions and cost of the system. When the inverter 1 valves are turned off, the contactors 207a and 207b always short-circuit to perform demagnetization. It can be safe. However, the constant transmission of current can create a braking force applied to the variable-flux motor 4.

Since there are two demagnetization units, that is, contactors 207a and 207b, it is possible to perform demagnetization even if one of them fails. This improves the reliability of the variable-flux motor drive system.

(Eleventh Embodiment)

FIG. 49 is a block diagram showing a variable-flux motor drive system according to an eleventh embodiment of the present invention. It differs from the tenth embodiment in that the contactor 207c is mounted between the wires of the variable-flux motor 4.

The switch 207c corresponds to the demagnetization unit of the present invention and performs demagnetization by shorting the wires of the variable magnet according to the demagnetization signal generated by the stop / demagnetize test unit 208c. In contrast to the tenth embodiment, the demagnetization unit of this embodiment shorts only one wire connection. This configuration can demagnetize the alternating magnet between the wires shorted by the contactor 207c, since the variable-flux motor 4 rotates.

FIG. 50 shows examples of a demagnetization unit capable of shorting the wires of a variable-flux motor 4 for demagnetization. According to an embodiment, the demagnetization unit, which achieves the demagnetization by short-circuiting, of at least one inter-wire connection of the variable-flux motor 4, consists of a contactor 207c shown in FIG. 50 (c). It can be a semiconductor switch. Instead of the contactor 207c, the demagnetization unit of the embodiment may be a combination of thyristors and non-conductive diodes shown in FIG. 50 (a). Alternatively, it can be a combination of self-extinguishing elements (such as GTO (shut-off valves), IGBT (insulated gate bipolar transistors) and MOSFET (metal-oxide-semiconductor junction field effect transistors)) and non-conducting elements, shown in FIG. 50 (b). The switches 207a and 207b of the tenth embodiment are interchangeable with demagnetization units using the aforementioned semiconductor switches. The rest of the configuration of the eleventh embodiment is similar to that of the tenth embodiment, and therefore, repeated explanations are omitted.

Then, the operation of the embodiment with the above configuration will be explained. The stop / demagnetize check unit 208c operates similar to the stop / demagnetize check unit 208b of the tenth embodiment. When the inverter 1 stops operation or when the variable-flux motor drive system needs to be protected, the stop / demagnetize check unit 208c determines that the variable magnet 53 should be demagnetized and generates a demagnetize signal that is output to the contactor 207c.

Based on the input demagnetization signal, the contactor 207c shorts the wires of the variable-flux motor 4 to demagnetize the variable magnet 53.

In addition to the results of the ninth and tenth embodiments, the variable-flux motor drive system of the eleventh embodiment has a contactor 207c serving as a demagnetization unit to perform demagnetization even after the inverter 1 is stopped. This embodiment requires only one load contactor, i.e., contactor 207c, which reduces costs.

(Twelfth Embodiment)

FIG. 51 is a block diagram showing a configuration of a variable flux motor drive system according to a twelfth embodiment of the present invention. This embodiment is different from the ninth embodiment shown in FIG. 39, in that the contactors 207a and 207b are installed between the wires of the variable-flux motor 4. According to this embodiment, the inverter 1 and the contactors 207a and 207b each correspond to a demagnetization unit of the present invention. The stop / demagnetize check unit 208d outputs the demagnetization flags FLG_DEMAG1 and FLG_DEMAG2 serving as demagnetize signals to the contactors 207a and 207b.

When the inverter 1 stops operation or when the variable-flux motor drive system is to be protected, the stop / demagnetize check unit 208d determines that the variable magnet 53 should be demagnetized and generate demagnetization signals. At this time, if any one of the inverter 1 and the contactors 207a and 207b corresponding to the demagnetize units is faulty, the stop / demagnetize check unit 208d generates a demagnetize signal to cause one reliable demagnetize unit to demagnetize. Like the tenth embodiment, the contactors 207a and 207b are subject to a demagnetize signal generated by the stop / demagnetize test unit 208d to demagnetize the variable magnet 53 by shorting its wires. The remaining configuration of the twelfth embodiment is the same as that of the ninth embodiment shown in FIG. 39, and therefore, repeated explanations are omitted.

Then, the operation of the twelfth embodiment with the above configuration will be explained. According to an embodiment, the protection signal PROT includes information related to whether or not each of the demagnetization blocks, i.e., the inverter 1 and the contactors 207a and 207b, is faulty and whether it is required to protect (failure state). According to the protection signal PROT, the stop / demagnetize test unit 208d checks each of the inverter 1 and the contactors 207a and 207b to determine which one is faulty.

FIG. 52 is a flowchart showing the operation of the stop / demagnetize check unit 208d according to an embodiment. First of all, the stop / demagnetize check unit 208d determines whether or not the variable-flux motor drive system should be protected (step S101). If system protection is not required, the stop / demagnetize check unit 208d determines whether or not the previous operation command is in the running state (RUN_CMD = 1) and indicates whether or not the current operation stop command (RUN_CMD = 0) (step S103). If these conditions are not satisfied, the work ends here.

If in step S103, the stop / demagnetize check unit 208d determines that the previous operation command is in the operation state and the current operation command sets the stop, normal demagnetization is performed (step S107). Normal demagnetization can be performed in any way. For example, the stop / demagnetize test unit 208d outputs the demagnetize signals to the switch 205b and the contactors 207a and 207b. Namely, block 208d outputs FLG_DEMAG1 = 1 and FLG_DEMAG2 = 1 to simultaneously perform demagnetization by inverter 1 and contactors 207a and 207b. The stop / demagnetize check unit 208d typically outputs the demagnetize signal only to switch 205b, i.e., FLG_DEMAG1 = 1, and FLG_DEMAG2 = 0 to force demagnetization to only inverter 1. This can reduce the number of open / close operations, reduce the failure rate and extend the timing of failures contactor services 207a and 207b.

If system protection is required in step S101, the stop / demagnetize check unit 208d determines whether the inverter 1 serving as the main inverter and the contactors 207a and 207b serving as auxiliary devices are faulty or reliable (step S109). If all of the main inverter 1 and auxiliary devices are reliable (did not fail), the stop / demagnetize check unit 208d generates demagnetize signals to perform normal demagnetization (step S107).

If one of the main inverter 1 and the auxiliary devices is not reliable (failed), the stop / demagnetize check unit 208d determines whether or not the inverter 1 is reliable (not failed) (step S111). If the inverter 1 is reliable, the stop / demagnetize check unit 208d generates a demagnetize signal (FLG_DEMAG1 = 1) to cause the inverter 1, which is reliable, to demagnetize, and outputs a signal to the switch 205b. As a result, as explained in the ninth embodiment, the switch 205b outputs 0. The inverter 1 shorts the wires of the variable-flux motor 4 to demagnetize the variable magnet 53 (step S113). Accordingly, the contactors 207a and 207b do not demagnetize.

If the inverter 1 is not reliable, but is faulty in step S111, the stop / demagnetize test unit 208d generates a demagnetize signal to cause the demagnetizers 207a and 207b to perform demagnetization and outputs a signal (FLG_DEMAG2 = 1) to the closures 207a and 207b. Like the second embodiment, the contactors 207a and 207b are subject to a demagnetize signal generated by the stop / demagnetize test unit 208d to shorten the wires of the variable magnet and demagnetize it (step S115). Accordingly, the inverter 1 does not perform demagnetization.

Thus, the variable-flux motor drive system of the twelfth embodiment provides, in addition to the results of the ninth to eleventh embodiments, the result that when any one of the demagnetization units is faulty, the stop / demagnetize test unit 208d generates a demagnetize signal, to force a reliable demagnetization unit to perform demagnetization. Namely, this embodiment can only use a reliable demagnetization unit to perform demagnetization without using a faulty demagnetization unit, thereby reliably protecting the system.

(Thirteenth Embodiment)

FIG. 53 is a block diagram showing a variable-flux motor drive system according to a thirteenth embodiment of the present invention. This embodiment differs from the tenth embodiment in that it further has a magnetization inverter 231, a current detection device 232, a magnetization current command processing unit 212, a voltage command processing unit 210a, and a PWM circuit 6a.

The magnetization inverter 231 corresponds to the magnetization unit of the present invention, is connected to a DC power supply 3, and applies the magnetization current to the magnetization winding of the variable-flux motor 4 to regulate the magnetic flux of the variable magnet 3 of the variable-flux motor 4. The magnetization inverter 231 also corresponds to the demagnetization unit of the present invention and uses the demagnetization signal generated by the stop / demagnetize test unit 208e to demagnetize the variable magnet 53. The current passed to the magnetization coil does not need to be regenerated and must be passed through in both positive and negative. directions for magnetization and demagnetization. Accordingly, the magnetization inverter 231 has a known configuration of a single-phase two-period bridge inverter. The current detection device 232 detects a magnetizing current passed to the magnetizing winding, and outputs the detected current to the voltage command processing unit 210a.

The magnetization current command processing unit 212 calculates the required magnetization current, generates a magnetization current command, and outputs the command to the voltage command processing unit 210a. Typically, the magnetization current is dependent on the magnetization history of the variable magnet 53. Accordingly, the magnetization current command processing unit 212, for example, has tabular information storing magnetization currents corresponding to the magnetization histories and the required magnetic flux values and calculates the required magnetization current according to the table information. The magnetization current must be accurately passed at high speed, and therefore PI control can be replaced by a hysteretic comparator.

The voltage command processing unit 210a resorts to an input magnetization current command to calculate and create a voltage command so that the magnetization current provided by the magnetization inverter 231 can be matched with the command. The voltage command is supplied to the PWM circuit 6a. The PWM circuit 6a uses an input voltage command to control on / off switching elements of the magnetization inverter 231.

Accordingly, the inverter 1, the magnetization inverter 231 and the contactor 207c of the embodiment each correspond to the demagnetization unit of the present invention.

With this configuration, the magnetic field generated by the magnetizing current directly controls the magnetization of the variable magnet 53, which is a permanent magnet with a low coercive force. Unlike the prior art, which transmits an excessively large current along the D axis to the main magnetizing winding, the embodiment uses a magnetizing current to magnetize to magnetically control the flux of the variable magnet 53.

This reduces the maximum permissible current of the inverter 1 in order to reduce the size, weight and cost of the inverter 1. The magnetizing winding can be integrated into the iron core 52 of the rotor, as shown in FIG. 30 and 31. In this case, the magnetic flux coupled to the magnetization winding in the rotor 51 does not change in time, and therefore the magnetization winding does not create a counter electromotive voltage. This helps to reduce the allowable load of the magnetization inverter 231.

When the inverter 1 is used for magnetization, the degree of freedom in the design of the inductance is small. The motor inductance is designed taking into account the output power and the efficiency of the motor, and therefore optimization for magnetization is not always given priority. On the other hand, the use of an exclusive use magnetizing winding increases the degree of freedom in designing the winding inductance and can optimize the inductance for magnetization.

The stop / demagnetize check unit 208e outputs the demagnetization flags FLG_DEMAG1, FLG_DEMAG2 and FLG_DEMAG3 serving as demagnetize signals to the magnetization current command processing unit 212, the current command processing unit 211 and the contactor 207c, respectively.

When the inverter 1 stops operation or when the variable-flux motor drive system is to be protected, the stop / demagnetize check unit 208e determines that the alternating magnet 53 should be demagnetized and generate demagnetization signals. At this time, if any one of the demagnetization units, i.e., the inverter 1, the magnetization inverter 231 and the contactor 207c is faulty, the stop / demagnetize test unit 208e generates demagnetization signals to cause one reliable demagnetization unit to perform demagnetization. Like the eleventh embodiment, the contactor 207c uses the demagnetize signal generated by the stop / demagnetize test unit 208e to short-circuit the wires of the variable magnet 53, thereby demagnetizing it. The rest of the configuration of the embodiment is the same as that of the tenth embodiment, and therefore, repeated explanations are omitted.

Then, an operation of an embodiment having the above configuration will be explained. According to an embodiment, the protection signal PROT contains information related to whether or not each of the demagnetization units must be protected (failure state), that is, inverter 1, magnetization inverter 231 and contactor 207c, for example, due to a failure. The stop / demagnetize test unit 208e uses the protection signal PROT to determine whether or not any of the inverter 1, the magnetization inverter 231, or the contactor 207c is in a failure state.

FIG. 54 is a flowchart showing the operation of the stop / demagnetize test unit 208e, according to an embodiment. First of all, the stop / demagnetize check unit 208c determines whether or not the variable-flux motor drive system should be protected (step S201). If system protection is not required, the stop / demagnetize check unit 208e determines whether or not the previous operation command is in the running state (RUN_CMD = 1) and indicates whether or not the current stop operation command (RUN_CMD = 0) (step S203). If these conditions are not satisfied, the work ends here.

If, in step S202, the stop / demagnetize check unit 208e determines that the previous operation command is in the operation state and the current operation command sets the stop, the magnetization inverter 231 serving as the magnetization circuit performs demagnetization (step S205). To cause the magnetization inverter 231 to demagnetize, the stop / demagnetize check unit 208e outputs a demagnetize signal, FLG_DEMAG1 = 1, to the magnetization current command processing unit 212. The magnetization current command processing unit 212 calculates the magnetization current necessary for the magnetization inverter 231 to perform demagnetization, and outputs the magnetization current command to the voltage command processing unit 210a. The voltage command processing unit 210a and the PWM circuit 6a operates as explained above. Magnetization inverter 231 passes magnetization current to demagnetize an alternating magnet.

If system protection is required in step S201, the stop / demagnetize check unit 208e determines whether the magnetization inverter 231 serving as the magnetization circuit is faulty or reliable (step S203). If the magnetization inverter 231 is reliable (did not fail), the stop / demagnetize check unit 208e outputs the demagnetize signal, FLG_DEMAG1 = 1, to the magnetization current command processing unit 212. Based on the demagnetization signal, the magnetization inverter 231 demagnetizes the variable magnet (step S205).

If the magnetization circuit is not reliable (failed), the stop / demagnetize check unit 208e determines whether or not the inverter 1 is reliable (or failed) (step S207). If the inverter 1 is reliable, the stop / demagnetize check unit 208e generates a demagnetize signal to cause the inverter 1, which is reliable, to demagnetize, and outputs FLG_DEMAG2 = 1 to the current command processing unit 211. The current command processing unit 211 calculates the current required for the inverter 1 to perform demagnetization, and outputs the current command to the voltage command processing unit 210. The current command processing unit 211 can also generate a current command so that the voltage command processing unit 210 can output 0. In this case, the inverter 1 shorts the wires of the variable-flux motor 4 like the ninth embodiment to demagnetize the variable magnet 53 (step S209).

If the inverter 1 is not reliable and failed in step S207, the stop / demagnetize test unit 208e generates a demagnetize signal (FLG_DEMAG3 = 1) to cause the contactor 207c, which is reliable, to perform the demagnetization, and outputs the signal to the contactor 207c. Like the eleventh embodiment, the contactor 207c uses the demagnetize signal generated by the stop / demagnetize test unit 208e to short-circuit the wires of the variable magnet 53 and demagnetize it (step S211).

Thus, the variable-flux motor drive system of this embodiment provides, like the twelfth embodiment, the result that when one or more of the demagnetize blocks is disabled, the stop / demagnetize check unit 208e generates demagnetize signals to cause a reliable demagnetize block perform demagnetization. Namely, this embodiment can only use reliable demagnetization units to perform demagnetization without using faulty demagnetization units, thereby reliably protecting the system.

The use of a magnetization inverter 231 can reduce the maximum permissible current of the inverter 1 in order to reduce the size, weight and cost of the inverter 1.

(Fourteenth Embodiment)

FIG. 55 is a block diagram showing a variable-flux motor drive system according to a fourteenth embodiment of the present invention. What is different from the ninth embodiment shown in FIG. 39, consists in that it has a DC voltage detection device 217 and an opposing electromotive voltage evaluation unit 219.

The DC voltage detecting device 217 corresponds to the first voltage detecting unit of the present embodiment and detects the DC voltage supplied from the direct current power supply 3 to the inverter 1. The counter electromotive voltage evaluation unit 219 uses the voltages and currents provided by the inverter 1 to evaluate the counter the electromotive voltage of the variable magnet 53 of the motor 4 with a variable magnetic flux. Only when the counter electromotive voltage estimated by the counter electromotive voltage estimator 219 is equal to or greater than the DC voltage detected by the DC voltage detecting device 217, the stop / demagnetize check unit 208f determines that the alternating magnet 53 needs to demagnetize and creates demagnetization signal.

Here, the evaluation of the counter electromotive voltage will be explained. When the inverter 1 is stopped (locked), the counteracting electromotive voltage can be realized by measuring the inter-wire voltage of the variable-flux motor 4. However, when the inverter 1 is operational, it is not possible to directly measure the counter electromotive voltage. According to an embodiment, when the inverter 1 shuts down or when the variable-flux motor drive system is stopped for protection, the stop / demagnetize check unit 208f determines that the variable magnet should be demagnetized and generates a demagnetize signal. For this, the counter electromotive voltage evaluation unit 219 evaluates the counter electromotive voltage during the operation of the inverter 1.

This will be explained on the known rotating coordinate system of the DQ axes. The axis D is defined in the same direction as the magnetic flux vector of the magnet. The steady state characteristic equation is as follows:

[Mat. 21]

Vd = R × Id - ω × Lq × Iq (28) Vq = R x Iq - ω × Ld × Id + E (29)

Here R is the winding resistance, Ld and Lq are the inductance along the D axis and inductance along the Q axis, Vd and Vq are the voltage along the D axis and voltage along the Q axis, respectively, Id and Iq are the current along the D axis and current along the Q axis, respectively, ω is the angular frequency of rotation (electric angle) of the rotor, and E is the counteracting electromotive voltage.

The estimated counter electromotive voltage Eh is expressed as follows:

[Mat. 22]

Eh = Vq - R × Iq - ω × Ld × Id (thirty)

With this expression, the counteracting electromotive voltage is computable.

The current detection device 2 detects the currents supplied to the variable-flux motor 4 and outputs the detected currents to the voltage command processing unit 210 and the counter electromotive voltage evaluation unit 219. Block 219 counteracting electromotive voltage converts the coordinates of the input currents of the phases U and W into currents on the D- and Q-axis, to obtain the current Id on the D-axis and current Iq on the Q-axis. In addition, the counter electromotive voltage estimator 219 uses the three-phase voltage commands provided by the switch 205c to perform a coordinate transformation that provides the D-axis voltage Vd and the Q-axis voltage Vq. Block 219 counteracting electromotive voltage can receive voltage along the D- and Q-axis by direct measurement of actual values. The rest of the configuration of this embodiment is the same as that of the ninth embodiment, and therefore, repeated explanations are omitted.

Then, the operation of the embodiment with the above configuration will be explained. While the inverter 1 is operating, the counter electromotive voltage estimator 219 estimates the counter electromotive voltage of the variable magnet 53 according to the voltages and currents provided by the inverter 1. The estimated counter electromotive voltage Eh based on equation (30) is located in the coordinates of the DQ axes. Since the voltage in the coordinates of the DQ axes is equal to the inter-wire voltage (RMS), the amplitude Eh ′ of the counter electromotive voltage converted to the inter-wire voltage is expressed as follows:

[Mat. 23]

The counter electromotive voltage evaluation unit 219 outputs the calculated Eh ′ to the stop / demagnetize test unit 208f. The DC voltage detecting device 217 detects the DC voltage Vdc supplied from the DC power source 3 to the inverter 1, and outputs the detected voltage to the stop / demagnetize test unit 208f.

When the inverter 1 stops operation or when the variable-flux motor drive system needs to be protected, the stop / demagnetize test unit 208f compares the counter electromotive voltage Eh ′ and the direct current voltage Vdc with each other and determines whether or not the alternating magnet 53 is demagnetized. Counter an electromotive voltage Eh ′ indicates a peak counteracting electromotive voltage. Accordingly, if the counter electromotive voltage Eh ′ is equal to or greater than the direct current voltage Vdc, the peak counter electromotive voltage has a probability of exceeding the direct current voltage Vdc. Accordingly, the stop / demagnetize check unit 208f determines that the variable magnet 53 should be demagnetized, generates a demagnetize signal (FLG_DEMAG = 1), and outputs the signal to switch 205c. If the opposing electromotive voltage Eh ′ is less than the direct current voltage Vdc, the stop / demagnetize test unit 208f does not generate a demagnetize signal. The rest of the work according to the embodiment is the same as that of the ninth embodiment, and therefore, repeated explanations are omitted.

As mentioned above, the variable-flux motor drive system of the embodiment provides, in addition to the result of the ninth embodiment, a result of comparing the counter electromotive voltage estimated by the counter electromotive voltage evaluating unit 219 with the DC voltage detected by the DC voltage detecting device 217, and determining whether or not demagnetization is to be performed. If the opposing electromotive voltage is less than the DC voltage of the DC power source 3, the embodiment does not perform demagnetization to reduce the number of times unnecessary demagnetization and extend the life of the elements.

If the counter electromotive voltage is equal to or greater than the DC voltage of the DC power supply 3, a large current will be passed to the inverter 1, which damages the elements and causes overheating. In addition, the variable flux motor 4 may receive braking force. To deal with them, the stop / demagnetize test unit 208f generates and outputs a demagnetize signal to prevent the inverter from receiving overvoltage and excessively high current, and to avoid braking force.

This embodiment determines stop or demagnetization by simply comparing the DC voltage with the opposing electromotive voltage. Electric trains, EV, HEV, ships and the like, use many drive systems to drive one object. In this case, each drive is not able to independently determine the speed (engine speed) of the object. In addition, an external force (wind, incline, and the like) acting on an object can accelerate the object. In this case, the determination of whether or not demagnetization is to be carried out can be made taking into account the maximum number of revolutions in the object’s motion schedule and the increase in the object’s speed (number of revolutions) due to the disturbing effect.

(Fifteenth Embodiment)

FIG. 56 is a block diagram showing a variable-flux motor drive system according to a fifteenth embodiment of the present invention. This embodiment is different from the ninth embodiment shown in FIG. 39, in that it comprises a voltage detection device 213, an overvoltage test unit 214, a timer 215, and an OR circuit 216.

The voltage detection device 213 corresponds to the second voltage detection unit of the present invention and detects the inter-wire voltage of the variable-flux motor 4.

If the inverter 1 is stopped and if the inter-wire voltage detected by the voltage detection device 213 is equal to or greater than a predetermined value, the overvoltage test unit 214 generates a demagnetization request signal and outputs it to the OR circuit 216.

The timer 215 corresponds to the timer block of the present invention and measures the time. When the inverter 1 is stopped, the timer generates a demagnetization request signal whenever a predetermined time passes and outputs the signal to the OR circuit 216.

The stop / demagnetize test unit 208g outputs the unlock control command Gst to the PWM circuit 6, the overvoltage test unit 214 and the timer 215. Accordingly, the overvoltage test unit 214 and the timer 215 are able to quickly sense whether inverter 1 is stopped or not according to the control command Gst unlocking.

When receiving the demagnetization request signal from one of the overvoltage test unit 214 and the timer 215, the OR circuit 216 outputs the demagnetization request signal to the stop / demagnetize test unit 208g.

The stop / demagnetize check unit 208g uses the demagnetize request signal generated by the overvoltage test unit 214 or the demagnetize request signal created by the timer 215 to determine whether or not the magnet is demagnetized, generates a demagnetize signal, and outputs the signal to switch 205d.

When the inverter 1 stops operation or when the variable-flux motor drive system is to be protected, the stop / demagnetize check unit 208g determines, like the ninth embodiment, that the variable magnet 53 is to be demagnetized and generates a demagnetize signal. The overvoltage test unit 214 and the timer 215 operate after the inverter 1 stops.

Although not shown in FIG. 56, a magnetic flux detection unit may be installed. The magnetic flux detection unit evaluates or detects the magnetic flux of the variable magnet 53, and generates a demagnetization request signal if the inverter 1 stops, and if the magnetic flux of the variable magnet is equal to or greater than a predetermined value. In this case, the stop / demagnetize check unit 208g uses the demagnetize request signal generated by the magnetic flux detection unit to determine whether or not the magnet magnet 53 should be demagnetized, create a demagnetize signal, and output the signal to switch 205d. The rest of the configuration is the same as that of the ninth embodiment, and therefore, repeated explanations are omitted.

Then, an operation of an embodiment having the above configuration will be explained. When the inverter 1 stops operation or when the variable-flux motor drive system is to be protected, the variable magnet 53 will be demagnetized. This operation is the same as in the ninth embodiment.

The stop / demagnetize test unit 208g outputs an unlock control command, Gst = 0, to stop the inverter 1. When the inverter 1 stops, the overvoltage test unit 214 checks to see if the wire-to-wire voltage detected by the voltage detection device 213 is equal to or greater than the predetermined value. The inter-wire voltage detected by the voltage detection device 213 is a counter electromotive voltage since the inverter 1 is stopped. Namely, the overvoltage test unit 214 carries out a check to determine whether the opposing electromotive voltage is equal to or greater than a predetermined value. The predefined value is freely set by the constructor or operator. Alternatively, it may be predefined in the overvoltage test block 214. If the inter-wire voltage (counter electromotive voltage) detected by the voltage detecting device 213 is equal to or greater than the predetermined voltage, the overvoltage test unit 214 generates a demagnetization request signal and outputs a signal to the OR circuit 216.

If the stop / demagnetize test unit 208g outputs an unlock control command, Gst = 0, inverter 1 stops operation. When the inverter 1 stops, the timer 215 measures the time, and whenever a predetermined time passes during the stop of the inverter 1, it generates a demagnetization request signal, which is output to the OR circuit 216. This predetermined time may be freely set by the designer or operator, or may be pre-set in timer 215.

By receiving a demagnetize request signal from any one of the overvoltage test unit 214 and a timer 215, the OR circuit 216 outputs a demagnetize request signal to the stop / demagnetize test unit 208g. Based on the demagnetization request signal generated by the overvoltage test unit 214 or the demagnetization request signal generated by the timer 215, the stop / demagnetize test unit 208g determines whether or not the magnet magnet 53 should be demagnetized, create a demagnetize signal, and output the signal to switch 205d. The rest of the work is the same as that of the ninth embodiment, and therefore, repeated explanations are omitted.

As mentioned above, the variable-flux motor drive system of the embodiment has an overvoltage test unit 214, and therefore can provide, in addition to the result of the ninth embodiment, the result of demagnetization, even after stopping the inverter 1, if the counter electromotive voltage is equal to or greater than a predetermined value to suppress an increase in counteracting electromotive voltage. In addition, the system has a timer 215, and therefore can demagnetize at predetermined intervals, even after stopping the inverter 1, to suppress an increase in the counteracting electromotive voltage. If the system is provided with the magnetic flux detection unit mentioned above, the system can perform demagnetization, even after stopping the inverter 1, if the magnetic flux of the alternating magnet 53 is equal to or greater than a predetermined value in order to suppress an increase in the counter electromotive voltage. As a result, this prevents the application of braking force and is a reliable protection of the system.

INDUSTRIAL APPLICABILITY

The variable magnetic flux motor drive systems of the present invention are applicable to drive electric train engines, electric vehicles, hybrid vehicles, and the like.

Claims (4)

1. A variable-flux motor drive system having an inverter for driving a variable-flux motor that uses a permanent magnet and a variable magnet, comprising: a torque command generating unit for generating torque commands for the variable-flux motor; a variable magnetic flux control unit for changing a magnetic flux according to a magnetizing current from an inverter and magnetizing an alternating magnet; a switch for transmitting predetermined current values along the D- and Q-axis based on a torque command from the torque command generation unit or magnetization current commands on the D- and Q-axis from the variable magnetic flux control unit; a magnetization request creation unit for creating, if the predetermined conditions are satisfied, a request to the variable magnetic flux control unit for magnetizing the variable magnet; and an unlock control command generation unit for generating unlock control commands for controlling the inverter according to the torque command based on the D- and Q-axis current or magnetization current commands on the D- and Q-axis from the switch. 2. A variable-flux motor drive system, comprising: a permanent magnet motor using a permanent magnet; an inverter for driving a permanent magnet motor; and magnetization means for transmitting a magnetizing current to control the magnetic flux of a permanent magnet, a permanent magnet at least partially having the ability of a variable magnet, whose magnetic induction is variable in response to the magnetization current from the inverter, magnetization means having the ability to transmit the magnetization current, which goes beyond the saturation zone of the magnetization of the magnetic material of a variable magnet. 3. A variable-flux motor drive system, comprising: a permanent magnet motor using a permanent magnet; an inverter for driving a permanent magnet motor; variable magnetic flux control means for transmitting a magnetizing current for controlling a magnetic flux of a permanent magnet; means for detecting the current of the permanent magnet motor; and magnetic flux estimating means for evaluating the magnitude of the magnetic flux from the voltage and current applied to the permanent magnet motor and the inductance of the winding serving as a parameter of the motor, a permanent magnet at least partially having the ability of a variable magnet, whose magnetic induction is variable according to the magnetization current from the inverter. 4. A variable-flux motor drive system having an inverter for driving a variable-flux motor, which has a permanent magnet and a variable magnet, comprising: a main control unit for controlling the inverter, so that the torque of the variable-flux motor corresponds to a torque command ; magnetizing winding for magnetizing an alternating magnet of a variable-flux motor; and a magnetization circuit for supplying a magnetizing current to a magnetizing winding.

Images