JP2009201284A - Variable speed driver of pm motor - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide a variable speed driver of a PM motor capable of performing pickup activation from a high-speed idling state by adding a few functions to the existing variable speed driver without superposing a high-frequency current or without flowing a pulse-like short circuit current. <P>SOLUTION: Upon picking up a motor when the motor in an idle state is started by being connected to the variable speed driver, current demands i<SB>d</SB><SP>*</SP>, i<SB>q</SB><SP>*</SP>of current control parts 1A, 1B are made zero, respectively, and in gain switching circuits 11, 12, 13, a gain in a current control system is made larger than that for control at normal time. In an arithmetic circuit 14, the rotation speed of the PM motor 1 is found from the time during zero-crossing of a detection current, and a position (phase) of the PM motor is found from the position of the zero-crossing and the rotation speed. The pickup activation is started with the found phase θ as the initial phase of coordinate conversion parts 2, 7, and afterward, control is switched to normal control. In the gain switching circuits, a gain on the "d" axis side in picking up the motor is made larger than that on the "q" axis side. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to a variable speed drive device for a PM (permanent magnet) motor, and more particularly to picking up and starting a motor from a high-speed idling state in a device including phase estimation means for estimating a rotor phase of a motor by an observer.

  The PM motor is controlled by using the magnetic saliency of the rotor of the motor, estimating the rotor phase (magnetic pole phase) with an observer, and controlling the motor speed and position based on this rotor phase. There is a position sensorless control method of the driving device.

  FIG. 4 is a block diagram of a PM motor position sensorless vector control device, which performs speed or torque control by magnetic pole position estimation.

The control system shown in FIG. 2 is provided with a torque current command i _q ^* and a field current command i _d ^* for the _d and _{q 2} axes by the speed control calculation means or the torque control calculation means. The deviation between the torque current command i _q ^* and the field current command i _d ^* and the detected current values i _q , i _d is proportionally integrated (PI) to calculate the voltage commands V _d , V for the d-axis and q-axis. _The current control units 1A and 1B for obtaining _q , and these voltage commands are rotated by the coordinate conversion unit 2 to convert them to fixed coordinates to obtain two-phase voltage commands Vα and Vβ, and the two-phase / three-phase conversion unit 3 further converts the three phases. _Are converted to voltage commands V _u , V _v , and V _w, and the armature current of the PM motor 5 is controlled by the inverse conversion circuit 4 based on these voltage commands.

The three-phase / two-phase converter 6 converts the detected currents i _u , i _v , i _w of the PM motor 5 into two-phase currents iα, iβ, and these currents are fixed by the coordinate converter 7 → rotating coordinates The detection currents i _d and i _q are obtained by conversion, and these are used as feedback currents to the current control units 1A and 1B.

The coordinate conversion units 2 and 7 perform coordinate conversion based on the magnetic pole phase θ of the PM motor, and instead of directly detecting the magnetic pole phase θ from the PM motor, the phase estimation unit 8 estimates the magnetic pole position. In this estimation calculation, the voltage commands Vα and Vβ, the current detection signals iα and iβ, and the angular velocity ω (motor rotational speed) are substituted into the voltage-current equation of the observer 8A having the motor model (R in the figure represents the motor). The estimated magnetic fluxes λα and λβ are obtained from the internal resistance of the model, L is the inductance), and the estimated phase θe is obtained from these by the arctangent calculation unit 8B. The differential operation unit 8C obtains the estimated angular velocity ω _e as the angular velocity ω by differentiating the estimated magnetic flux θe.

  By the way, in a device (variable speed drive device) that drives a motor with an inverter (inverse conversion circuit), an instantaneous power failure occurs in the power supply of the inverter, the inverter is temporarily stopped, and the inverter is connected to the idling motor after power recovery There is a method of picking up the motor in which the motor is accelerated from the current speed to the set speed.

  When picking up the motor, if the output of the inverter is connected to the motor with zero voltage, the idling PM motor generates a speed electromotive force at the terminal. Therefore, when the idling speed is high, the PM motor terminal is short-circuited. Generates excessive current.

In order to suppress this, it is necessary to output from the inverter a voltage having the same amplitude and the same phase as the speed-induced electromotive force of the idling PM motor, but the rotor position of the motor is estimated by an observer as shown in FIG. In a position sensorless control device, the detected currents I _u , I _v , and I _w are influenced by the induced electromotive force of the motor, so that the rotor position cannot be estimated. As a result, a voltage that matches the rotor position and speed of the motor Amplitude and phase cannot be controlled as inverter output.

As a method for performing position sensorless detection of a PM motor, there is a method for obtaining a rotor position of a motor from feedback currents iα and iβ when a high-frequency current is superimposed on motor current commands i _d ^* and i _q ^* . When this rotor position estimation method is used for starting the motor from the idling state, if the control gain of the rotor position estimation means is low, current control becomes unstable, and undesired torque is generated. The overcurrent protection operation makes it impossible to restart. Therefore, there is also an activation method that avoids overcurrent by increasing the control gain of the rotor position estimation by the rotor position estimation means in this activation method to speed up the response (for example, see Patent Document 1).

  Another starting method is to use the principle that the output of the inverter is set in a short-circuit state for a short time, and then the pulsed short-circuit current generated from the PM motor is generated in exactly the opposite phase of the induced electromotive force. This short-circuit current generation is executed twice with a time interval, and there is an activation method in which the speed is estimated from the time difference and the phase difference, and further the position (phase) is estimated.

Since this starting method requires that the short-circuit current generation interval of the rotation of the rotor be within 180 ° in terms of electrical angle, as an estimation method of speed / position (phase) without these restrictions, There is also a method in which the short-circuit current is generated three times or more, the time difference and phase difference of these generated currents are obtained, and the speed and phase are obtained by taking the difference (see, for example, Patent Document 2).
JP 2005-80458 A JP 2004-215466 A

  In the starting method of the above-mentioned patent document 1, the rotor position estimating means estimates the rotor position of the motor from the currents iα and iβ when the high frequency current is superimposed on the current command, and increases the control gain at the time of starting to respond. Overcurrent is avoided by speeding up. However, the high-frequency current superimposition control has much worse noise (generation of magnetic noise due to high frequency) and efficiency than the observer control (high frequency is a reactive power component that does not contribute to torque). Incorporating a high-frequency current superimposing circuit and an arithmetic algorithm complicates the system and increases the load on the CPU that performs the arithmetic operation. In particular, the arithmetic processing speed may be a problem in a high-speed rotation region.

  As in Patent Document 2, the start-up method using an observer that estimates the rotor position of a motor from a pulse-like short-circuit current has a function of performing a pick-up that estimates the position from a pulse-like short-circuit current. There is a problem that a significant improvement of the apparatus is required to add the pick-up activation function.

  The object of the present invention is to provide a variable speed drive device for a PM motor that can be picked up and started from a high speed idling state by adding a little function to an existing position sensorless control device without flowing a high frequency current superposition or a pulsed short circuit current. It is to provide.

  The present invention increases the gain of the current control system at the time of picking up the motor and increases the responsiveness of the current control system so that the above problem can be picked up even in a high speed idling state where the induced voltage of the motor is high. The absolute value (crest value) of the motor current generated from the induced voltage of the motor is suppressed, and the motor speed and position (phase θ) are obtained from the zero cross of the motor current at this time by calculation. The output of the combined inverse conversion circuit is obtained and has the following configuration.

(1) A current control system for controlling the output voltage of the inverse conversion circuit based on the feedback control of the PM motor current command and detection current, the PM motor α-axis and β-axis voltage commands, and the α-axis and β-axis In a variable speed drive device for a PM motor comprising phase estimation means for estimating the phase of a rotor by an observer from a detected current,
A pick-up gain circuit that sets the current command of the current control system to 0 and picks up the gain of the current control system larger than the normal control when picking up the motor that starts the idling motor;
Calculation means for obtaining the rotational speed of the PM motor from the time between the zero crosses of the detected current, and obtaining the position (phase) of the PM motor from the position and rotational speed of the zero cross;
It is provided with.

(2) The pick-up gain circuit is
A gain setter that sets the gain when picking up the motor with a coefficient relative to the gain at the normal time,
A multiplier that multiplies the voltage commands Vd and Vq of the d-axis and q-axis of the current control system by a gain set by the gain setting device when picking up the motor;
Is provided with a gain switching circuit that makes the gain at the time of picking up the motor of the current control system larger than the control at the normal time.

  (3) The gain switching circuit is characterized in that the gain on the d-axis side when picking up the motor is larger than that on the q-axis side.

  As described above, according to the present invention, the absolute value (crest value) of the motor current generated from the induced voltage of the motor is increased by increasing the gain of the current control system and increasing the response of the current control system when picking up the motor. The motor speed and position (phase θ) is calculated from the zero cross of the motor current at this time, and the output of the inverse conversion circuit that matches the motor speed and phase is obtained. Without adding a short-circuit current, it can be picked up and started from a high-speed idling state by adding a few functions to the existing variable-speed drive device.

FIG. 1 is an overall configuration diagram of a variable speed drive device for a PM motor showing an embodiment of the present invention. 4 differs from FIG. 4 in that a gain setting circuit 11 and a coefficient are used as a pick-up gain circuit to multiply the PI calculation result of the voltage commands V _d and V _q by the current control units 1A and 1B when picking up the motor. A gain switching circuit comprising the units 12 and 13 is provided. In addition, an arithmetic circuit 14 is provided that obtains the rotational speed of the PM motor from the time between the zero crosses of the detected currents iα and iβ, and obtains the position (phase) of the PM motor from the position and rotational speed of the zero cross. When the motor is picked up, the variable speed drive device is in a regenerative mode, and a power storage device such as a battery or a capacitor (not shown) or a power generation braking resistor that absorbs the regenerative energy is required.

  By adding a gain switching circuit, the response of the current control system is increased by increasing the gain of the current control system when picking up the motor, and the absolute value of the motor current generated from the induced voltage of the motor by this highly responsive current control ( (Crest value) is suppressed. The arithmetic circuit 14 calculates the motor speed and position (phase θ) from the zero cross of the motor current at this time, and initializes the phase θ in the coordinate conversion units 2 and 7. With this setting, the variable speed drive device obtains the output of the inverse conversion circuit 4 in accordance with the speed and phase of the motor, picks up and activates, and then returns to normal control.

In the configuration of FIG. 1, in the normal state, the magnetic pole phase θ is estimated by the phase estimation unit 8 configured by the observer 8A and the like, and position sensorless control is performed based on this estimated value. At the time of motor pick-up control, the current commands I _d ^* and I _q ^* are both 0, and the phase θ information is not known, so it is 0. At this time, since the detected current I _d is synchronized with the induced voltage of the motor, the rotational speed of the PM motor is obtained by measuring the time between zero crossings of the detected current I _d . Further, the position (phase) of the PM motor is obtained by using the zero cross position (point) and the rotation speed of the current I _d .

  The rotation speed and position calculation circuit 14, the gain setting unit 11, and the coefficient units 12 and 13 are added as dedicated digital calculation means, or added as a software configuration when the control device is configured as a computer. Can do. In either case, only a few calculations and control functions are required, and no significant device change or algorithm addition is required.

  The gain setting by the gain setting unit 11 vibrates when the gain is increased in the normal control. Therefore, the gain must be properly used for picking up and normal. Here, the gain at the normal time is set to 1, and the gain at the time of picking up is made larger than 1.

  FIG. 2 shows a specific example of the gain switching circuit. The gain switching determination unit 15 determines whether the position sensorless control device is in the normal control mode or the pickup control mode, and switches the changeover switch 16 to one of the normal gain and the pickup gain.

In this figure, the gain applied to V _d and V _q is shown as being the same, but in practice, it is more effective to increase the gain applied to the V _d side. Further, the gain to be actually applied may be about 2 to 8 times when the normal time is 1. If it is too large, problems such as overflow will occur.

  In this way, picking up can be easily performed by properly using the gain. FIG. 3 shows an image of the current waveform when the gain is switched, the solid line shows the waveform when the gain is not operated, and the broken line shows the waveform when the pick-up gain is multiplied. The current at the normal time and the pick-up time The frequency does not change, and only the current peak value changes when the gain is switched. Note that the regenerative energy generated by the induced voltage during idling of the motor may be charged to a power storage device such as a battery or a capacitor (not shown) or consumed by a power generation braking resistor.

1 is an overall configuration diagram of a variable speed drive device for a PM motor showing an embodiment of the present invention. A specific example of a gain switching circuit. Current waveform image when switching gain. The structural example of the variable speed drive device of PM motor.

Explanation of symbols

1A, 1B Current control unit 2 Coordinate conversion unit 3 2-phase / 3-phase conversion unit 4 Inverse conversion circuit 5 PM motor 6 3-phase / 2-phase conversion unit 7 Coordinate conversion unit 8 Phase estimation unit 8A Observer 8B Inverse tangent calculation unit 8C Differentiation Operation unit 11 Gain setting unit 12, 13 multiplier
14 Arithmetic circuit 15 Gain judgment unit 16 Changeover switch

Claims (3)

From the current control system that controls the output voltage of the inverse conversion circuit based on the feedback control of the PM motor current command and detected current, the PM motor α-axis and β-axis voltage commands, and the α-axis and β-axis detected currents In a variable speed drive device for a PM motor comprising phase estimation means for estimating the phase of a rotor by an observer,
A pick-up gain circuit that sets the current command of the current control system to 0 and picks up the gain of the current control system larger than the normal control when picking up the motor that starts the idling motor;
Calculation means for obtaining the rotational speed of the PM motor from the time between the zero crosses of the detected current, and obtaining the position (phase) of the PM motor from the position and rotational speed of the zero cross;
A variable speed drive device for a PM motor, comprising: The pick-up gain circuit is
A gain setter that sets the gain when picking up the motor with a coefficient relative to the gain at the normal time,
A multiplier that multiplies the voltage commands Vd and Vq of the d-axis and q-axis of the current control system by a gain set by the gain setting device when picking up the motor;
2. The variable speed drive device for a PM motor according to claim 1, further comprising a gain switching circuit for increasing a gain at the time of picking up the motor of the current control system larger than that at a normal control.   The variable speed drive device for a PM motor according to claim 2, wherein the gain switching circuit has a configuration in which a gain on the d-axis side when picking up the motor is larger than that on the q-axis side.

Images