JP2017077122A - Motor control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain a motor control device capable of detecting step-out of an observer that estimates a rotational speed of a motor and a magnetic pole position with high accuracy, in a case where a speed of a permanent magnet synchronous motor is controlled at a high response.SOLUTION: A motor control device comprises: a first estimation unit 6a that calculates an estimated speed of a permanent magnet synchronous motor 300 and an estimated magnetic pole position by using an adaptive identification unit 6a2 and an observer 6a1 based on a motor model; a second estimation unit 6b that calculates an estimated speed and an estimated magnetic pole position of the permanent magnet synchronous motor 300 from a motor induction voltage; and a determination and start processing unit 6c that performs reset determination of the estimated speed and the estimated magnetic pole position of the first estimation unit 6a by using the estimated magnetic pole position of the first estimation unit 6a and the estimated magnetic pole position of the second estimation unit 6b, or the estimated speed of the second estimation unit 6b.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a motor control device that controls a permanent magnet synchronous motor by performing sensorless control without using rotor position information and rotor speed information of the motor.

  The position control of the permanent magnet synchronous motor requires the rotor position information of the motor. Similarly, the speed control of the permanent magnet synchronous motor requires the rotor speed information of the motor. In the current control of the permanent magnet synchronous motor, the magnetic pole position of the motor is calculated from the position information of the rotor, and a coordinate conversion process for converting an alternating current signal into a direct current signal is performed using this. In order to acquire this position information and speed information, it is necessary to connect a position speed sensor such as an incremental encoder, absolute encoder, or resolver to the rotor of the motor. However, the use of position and speed sensors has problems such as increased cost, increased installation space, and sensor failure. Control methods that estimate motor rotor speed and magnetic pole position from motor voltage and current information are widely used. Yes. A sensor that detects position information and speed information is called a position speed sensor, and a control system that estimates the rotor speed and magnetic pole position is called sensorless control.

  As described above, this sensorless control is mainly used as an alternative to the position / speed sensor, but can be applied to a position / speed sensor failure detection by applying it to a control device for a motor with a position / speed sensor. Further, it is possible to realize an operation of performing a motor control process using the estimated speed and the estimated magnetic pole position as an alternative to the magnetic pole position information and the speed information from the position / velocity sensor after the failure is detected, and performing a motor stop process according to a predetermined speed command.

  Next, an outline of the sensorless control method will be described. The methods applicable to the permanent magnet synchronous motor are roughly classified into two types: an observer method and an induced voltage method. First, the observer method will be described. A mathematical model of a permanent magnet synchronous motor is provided inside the motor control device, and a voltage value applied to the permanent magnet synchronous motor is input to the mathematical model to estimate the current of the permanent magnet synchronous motor. In parallel with the model calculation, a motor current is detected using a current sensor, and the detected motor current is compared with the estimated current. Assuming that the difference in current is caused by an error in the motor speed information included in the mathematical model, the estimated speed is obtained by performing feedback correction of the estimated speed included in the model. The estimated magnetic pole position required for current control can be estimated by integrating the estimated speed. This is a method for obtaining an estimated speed and an estimated magnetic pole position based on so-called adaptive identification. In many cases, the mathematical model is not only a mathematical model of a motor but also an observer method in which a signal obtained by multiplying a difference between an estimated current and a detected current by a predetermined gain is added to a voltage value input to the mathematical model. When this observer method is used, there are merits such that the signal response of the estimated current can be set to a desired value, and robustness against the error of the motor constant is provided.

Next, the induced voltage method will be described. This is based on the induced voltage of the permanent magnet synchronous motor. The induced voltage is generated by the temporal change in the magnetic flux of the permanent magnet synchronous motor rotor, and is calculated from the motor current and motor voltage based on the motor model. it can. In the case of a three-phase permanent magnet synchronous motor, a three-phase sinusoidal induced voltage waveform is obtained with respect to the magnetic pole position, and the magnetic pole position can be calculated by arctangent calculation.

Both of these sensorless control methods can estimate the magnetic pole position and motor speed of the motor. Both methods use voltage information applied to the motor during the calculation process. At this time, the observer method has a relatively low sensitivity of the high-frequency voltage disturbance flowing into the sensorless control system included in the voltage information, and can obtain a relatively smooth estimated speed and estimated magnetic pole position. On the other hand, the induced voltage method has very high sensitivity to high-frequency voltage disturbance, and a high-frequency voltage disturbance component may flow into the estimated speed or the estimated magnetic pole position to cause a transient estimation error. Therefore, when the permanent magnet synchronous motor is controlled by sensorless control, the observer method is often adopted rather than the induced voltage method. In the stop process at the time of position / speed sensor failure, there is a case where it is desired to stop the permanent magnet synchronous motor as soon as possible, but there is also a case where it is desired to stop along a predetermined speed command locus due to the convenience of the load device connected to the motor. Under such conditions, the observer method is particularly employed. The high-frequency voltage disturbance described above includes, for example, a PWM carrier ripple component included in the inverter output voltage, a dead time voltage, and an induced voltage harmonic possessed by the permanent magnet synchronous motor.
Another difference between the two sensorless controls is the presence or absence of the following step-out. The observer method has a structure in which the estimated speed information is corrected by current estimation error feedback, so that there is a delay in speed estimation. When this delay is significant, a deviation occurs between the estimated magnetic pole position and the permanent magnet synchronous motor magnetic pole position. When this deviation increases, a state called out-of-step where the estimated magnetic pole position and the magnetic pole position of the permanent magnet synchronous motor are not synchronized is brought about. In this step-out state, it is difficult for the observer method to return to the non-step-out state. However, the induced voltage method does not have a feedback structure, so even if an instantaneous magnetic pole position estimation error occurs when the permanent magnet synchronous motor magnetic pole position or the motor speed changes suddenly, it returns to the non-step-out state unlike the observer method. Is possible.

  A case will be described in which a position / speed sensor failure detection and subsequent motor stop processing are performed by applying the above sensorless control method. In the permanent magnet synchronous motor, the above two types of sensorless control methods can be implemented. However, in order to realize reliable position / speed sensor failure detection and stable motor stop control due to the influence of the high-frequency voltage disturbance, as much as possible. It is desirable to use the estimated speed and the estimated magnetic pole position by the observer method. Therefore, when the observer method is operated, the following phenomenon occurs. When the position control or speed control of a permanent magnet synchronous motor is performed with high response by control with a position speed sensor, the observer method based on adaptive identification cannot follow changes in speed and position and can easily be out of step. . As described above, in the observer method, once the step is out, it is difficult to return to the original synchronized state. In the out-of-step state, accurate speed estimation and position estimation cannot be performed, so that failure detection of the position / speed sensor cannot be performed. For such a problem, a technique disclosed in Patent Document 1 is disclosed. The technique shown in Patent Document 1 uses the first estimated speed calculated from the motor induced voltage obtained from the motor voltage and the motor current, and the second estimated using the PLL so that the motor estimated magnetic pole position follows the true magnetic pole position. Two types of estimated speed are derived. That is, step-out detection is performed by comparing two types of estimated speeds with different calculation methods.

Japanese Patent No. 4198162

When applying observer-based sensorless control for the purpose of detecting the failure of the position / velocity sensor and performing stable motor stop control processing when the failure occurs, the technique described in Patent Document 1 described above is used and the PLL described in Patent Document 1 is used. The object can be achieved by replacing the estimation unit with a first estimation unit using an observer. However, when the step-out detection is performed by comparing the first estimated speed by the observer and the second estimated speed by the induced voltage, the following problems occur when the motor performs variable speed operation.
In general, the first estimated speed based on the observer method has a specific delay transmission characteristic with respect to the motor speed. On the other hand, since the second estimated speed by the induced voltage method is obtained by obtaining the estimated magnetic pole position from the induced voltage waveform of the permanent magnet synchronous motor and differentiating the estimated magnetic pole position, the first estimated speed is the motor speed. The transfer characteristics for are different. This difference in transfer characteristics is particularly noticeable in the high frequency region. “Frequency” here is a component contained in the motor speed. When the motor is driven at a high acceleration or when frequent acceleration and deceleration are performed, the motor is driven at a high jerk (time change rate of acceleration). When driven, the motor speed contains many high frequency components. At this time, a transient difference occurs between the first estimated speed and the second estimated speed even if the observer does not step out (also referred to as observer step out). When detecting an observer step-out based on the difference between the first and second estimated speeds, step-out detection criteria such as the magnitude of the difference between the estimated speeds and the duration thereof are provided. Because of the principle that a transient difference occurs, the locus of the estimated speed difference differs depending on the locus of the motor speed, and thus there is a problem that requires labor to set the detection reference for the step-out. Further, even if the out-of-step detection reference is adjusted with a predetermined motor speed trajectory, it changes with a different motor speed trajectory, and thus observer out-of-step may be erroneously detected. In the induced voltage method, the estimated magnetic pole position is obtained from the induced voltage waveform of the permanent magnet synchronous motor, and the estimated speed is obtained by differentiating the estimated magnetic pole position. Originally, the induced voltage waveform is easily influenced by high-frequency voltage disturbance and becomes a signal including noise. Further, since the high frequency component is amplified by the differential processing, when the first estimated speed and the second estimated speed are simply compared, in this case, too, the step-out detection is erroneously performed despite the non-step-out state due to noise. There was a case.

  The present invention has been made to solve the above-described problems. When the speed of the permanent magnet synchronous motor is controlled at a high response, the observer is removed to estimate the motor rotation speed and the magnetic pole position. An object of the present invention is to obtain a motor control device capable of detecting a tone with high accuracy.

  The motor control device according to the present invention is a motor control device including an estimator for estimating the rotor speed and the magnetic pole position of a permanent magnet synchronous motor, and the estimator is based on an observer and an adaptive identifier based on a motor model. A first estimation unit for obtaining an estimated speed and an estimated magnetic pole position of the permanent magnet synchronous motor, a second estimation unit for obtaining an estimated speed and an estimated magnetic pole position of the permanent magnet synchronous motor from the motor induced voltage, and an estimation of the first estimation unit A determination / startup processing unit for determining whether to reset the estimated speed of the first estimating unit and the estimated magnetic pole position using the estimated magnetic pole position and the estimated magnetic pole position of the second estimating unit or the estimated speed of the second estimating unit. is there.

  According to the motor control device of the present invention, transient speed estimation is included in the difference between the first estimated speed from the first estimating section that is the observer method and the second estimated speed from the second estimating section that is the induced voltage method. The content of high-frequency components that cause errors can be suppressed, and the accuracy of step-out detection can be improved.

It is a block diagram which shows the motor control apparatus in Embodiment 1 of this invention. It is a flowchart for demonstrating the flow of a process of the determination and starting process part in the motor control apparatus which concerns on Embodiment 1 of this invention. It is the figure which arranged and represented the operation | movement in the estimation part in the motor control apparatus which concerns on Embodiment 1 of this invention. It is a block diagram which shows the detailed structure of the magnetic pole position and speed estimator in the motor control apparatus which concerns on Embodiment 1 of this invention. It is explanatory drawing explaining operation | movement of the motor control apparatus in Embodiment 1 of this invention. It is explanatory drawing which showed the example of each signal waveform at the time of the step-out of the 1st estimation part in the motor control apparatus which concerns on Embodiment 1 of this invention. It is a figure for demonstrating operation | movement of the motor control apparatus in Embodiment 2 of this invention. It is a block diagram which shows the motor control apparatus in Embodiment 3 of this invention.

Embodiment 1 FIG.
FIG. 1 is a block diagram for explaining Embodiment 1 of the present invention, and shows a control device for a permanent magnet synchronous motor. The motor control device shown in FIG. 1 controls the speed of the permanent magnet synchronous motor based on the position / speed sensor information, and performs failure detection of the position / speed sensor in parallel. Furthermore, the structure of the control apparatus of a permanent magnet synchronous motor in the case of continuing the speed control of a permanent magnet synchronous motor based on sensorless control after failure detection is shown. FIG. 1 shows not only a control device for a permanent magnet synchronous motor but also a permanent magnet synchronous motor, a position speed sensor, and a host controller for generating a speed command.

As shown in FIG. 1, a permanent magnet synchronous motor control device 100 includes a speed controller 1, a current controller 2, an inverter 3, a current sensor 4, a signal switch 5, a magnetic pole position / speed estimator 6, and a speed converter 7. , A magnetic pole position converter 8 and a failure detector 9.
The current controller 2 includes a current controller 2a, coordinate converters 2b and 2c, and a PWM processor 2d. The magnetic pole position / speed estimator 6 further includes a first estimation unit 6a, a second estimation unit 6b, a determination / startup processing unit 6c, and a mechanical speed conversion unit 6d.

Next, the operation of each unit will be described. The speed controller 1 executes speed control processing based on the speed command 10 (ω ^* ) and the feedback speed 15 (ω) transmitted from the host controller 200, and outputs current commands 11 (Id ^* , Iq ^* ). . The current controller 2 inputs a current command 11 (Id ^* , Iq ^* ), a coordinate conversion phase 16 (θ), and a detected current 13 (Iu, Iv, Iw), executes control processing, and outputs a switching command 14 To do. The inverter 3 operates according to the switching command 14 and supplies power to the permanent magnet synchronous motor 300. The current flowing through the permanent magnet synchronous motor 300 is detected by the current sensor 4 and sent to the current controller 2 as the detected current 13 (Iu, Iv, Iw). Inside the current controller 2, the current controller 2a inputs the current command 11 (Id ^* , Iq ^* ) and the detected current (Id, Iq), performs control processing so that the difference between the two becomes zero, and performs the voltage command (Vd ^* , Vq ^* ) is output. The voltage command (Vd ^* , Vq ^* ) and the current command 11 (Id ^* , Iq ^*) are signals on the rotational coordinates synchronized with the magnetic pole position of the motor rotor, and are expressed on the three-phase stationary coordinates by the coordinate converter 2b. Or converted to a signal on the rotating coordinates. The detected current 13 (Iu, Iv, Iw) obtained from the current sensor 4 is converted into currents Id and Iq on the rotational coordinates by the coordinate converter 2c. The voltage command (Vd ^* , Vq ^* ) is converted into a voltage command 12 (Vu ^* , Vv ^* , Vw ^* ) on a stationary coordinate by the coordinate converter 2b. This coordinate conversion process uses the coordinate conversion phase 16 (θ). This coordinate conversion phase 16 (θ) corresponds to the magnetic pole position of the permanent magnet synchronous motor. The motor position signal 17 output from the position speed sensor 400 is the mechanical position of the rotor of the permanent magnet synchronous motor 300. The magnetic pole position θdet rotates by the number of pole pairs of the motor with respect to one rotation of the machine position. The magnetic pole position converter 8 performs the above processing and outputs the detected magnetic pole position 19 (θdet) of the permanent magnet synchronous motor 300. The speed converter 7 differentiates the motor position signal 17 and outputs a detected speed 18 (ωdet). Since the operations and configurations of the speed controller 1 and the current controller 2 described above are well-known techniques, a detailed description thereof will be omitted.

Next, operations of the failure detector 9 and the signal switch 5 will be described. During a normal operation in which the position / velocity sensor 400 is not malfunctioning, the detected magnetic pole position 19 (θdet) is assigned to the coordinate conversion phase 16 (θ) via the switching unit 5b, and the detection speed 18 (ωdet) is assigned to the switching unit 5a. , The speed controller 1 and the current controller 2 are operated. When the failure detector 9 detects a position / velocity sensor failure and the sensor is detected as a failure, a failure detection signal 22 is issued, and the signal switch 5 switches the switching units 5a and 5b accordingly. Change the signal assignment. When the position / speed sensor 400 fails, the estimated magnetic pole position 20 (θest) is assigned to the coordinate conversion phase 16 (θ) via the switching unit 5b, and the estimated speed 21 (ωest) is assigned to the feedback speed 15 (ω).
With such a configuration, it is possible to continue the operation of the permanent magnet synchronous motor 300 even when the position / speed sensor 400 fails. The failure detection signal 22 generated by the failure detector 9 is also transmitted to the host controller 200, and a speed command pattern for stopping the motor is generated as the speed command 10. Fault detection of the position / speed sensor 400 is performed by comparing the detected magnetic pole position 19 (θdet) with the estimated magnetic pole position 20 (θest). For example, there is a method of determining a failure when the difference between the two is a predetermined value or more and the state continues for a predetermined time or more. In addition, there are various methods of failure detection depending on the failure mode of the position / speed sensor 400. For example, the abnormality detection of the internal logic circuit is performed based on the consistency of the logic signal in the position / speed sensor, or the cable is detected from the impedance change. And disconnection detection. The failure detector 9 performs the above processing group together to detect a failure of the position / speed sensor 400. Since the operations of the failure detector 9 and the signal switching unit 5 described above are well-known techniques as application examples of the sensorless control technique, detailed description thereof is omitted.

  Next, details of the configuration and operation of the magnetic pole position / velocity estimator 6 will be described. The magnetic pole position / speed estimator 6 receives the voltage command 12 and the detection current 13 and outputs an estimated magnetic pole position 20 (θest) and an estimated speed 21 (ωest). The magnetic pole position / velocity estimator 6 mainly includes a first estimation unit 6a, a second estimation unit 6b, and a determination / activation processing unit 6c. The first estimation unit 6a performs a magnetic pole position / speed estimation operation by adaptive identification using an observer, and outputs a first estimated magnetic pole position 6e and a first estimated speed 6f. The second estimation unit 6b performs a magnetic pole position / speed estimation operation using an induced voltage generated by a permanent magnet, and outputs a second estimated magnetic pole position 6g and a second estimated speed 6h. The determination / activation processing unit 6c is configured so that the first estimation unit 6a can be activated stably, and can be returned using the information from the second estimation unit 6b even if the first estimation unit 6a is out of step. The determination operation is performed according to the flowchart shown in FIG. Details of the determination operation will be described later. Since both the first estimated speed 6f and the second estimated speed 6h are the rotation speeds of the magnetic pole positions of the permanent magnet synchronous motor, the estimated machine speed is converted by dividing the number of pole pairs of the permanent magnet synchronous motor by the machine speed conversion unit 6d. Output as estimated speed 21 (ωest).

When the motor controller is started assuming that there is no abnormality in the position speed sensor 400 connected to the permanent magnet synchronous motor 300, the determination / startup processing unit 6c inputs the second estimated speed 6h from the second estimating unit 6b. It is determined whether or not the absolute value of the estimated speed is less than a predetermined value ωth that is a threshold value (step S201), and if it is less than the predetermined value ωth, a reset instruction signal 6i is output to reset the first estimating unit 6a (step S202). . When the first estimation unit 6a receives the reset instruction signal 6i and needs to be reset, the first estimation magnetic pole position 6e calculated internally is overwritten with the second estimation magnetic pole position 6g.
Further, the first estimated speed 6f is overwritten with the second estimated speed 6h. While the reset instruction signal 6i is generated, the first estimating unit 6a continues the reset process. At this time, the first estimated magnetic pole position 6e and the first estimated speed 6f are overwritten with the information from the second estimation unit 6b, so that the observer and the adaptive identifier in the first estimation unit 6a are stopped.
Further, the estimated magnetic pole position 20 (θest) is equal to the second estimated magnetic pole position 6g. The estimated speed 21 (ωest) is equal to the second estimated speed 6h. If the absolute value of the second estimated speed 6h is less than the predetermined value (threshold value) ωth, the reset process is continued (step S202). If a reset instruction is issued in step S202, it is determined in step S206 whether or not the motor operation has ended. If the motor operation has not ended, the process returns to step S201, and if the motor operation has ended. The process ends.

When the permanent magnet synchronous motor 300 is appropriately controlled by feedback of the position speed sensor information, and the motor speed increases and the absolute value of the second estimated speed 6h becomes the predetermined value ωth, the reset instruction is stopped. From this time, the first estimation unit 6a starts the magnetic pole position / velocity estimation operation using the observer or adaptive identifier originally provided. The second estimated speed 6h at the moment when the first estimating unit 6a starts to operate is set as the initial value of the first estimated speed 6f. The operation of the observer and adaptive identifier 6a2 is started with the second estimated magnetic pole position 6g at the same moment as the initial value of the first estimated magnetic pole position 6e. Thereby, a 1st estimation part can be started stably.
Further, the difference between the first estimated magnetic pole position 6e and the second estimated magnetic pole position 6g is monitored to determine whether or not the absolute value θer of the difference between the two is greater than or equal to a predetermined threshold θerth (step S203). Is greater than or equal to the threshold θerth and continues for the time Tth or longer (step S204), the determination / startup processing unit 6c outputs the reset instruction signal 6i, assuming that the observer of the first estimation unit 6a has stepped out. (Step S205). If a reset instruction is issued in step S205, it is determined in step S206 whether or not the motor operation has ended. If the motor operation has not ended, the process returns to step S201, and if the motor operation has ended, the process is performed. finish.

  The determination / activation processing unit 6c is not limited to the difference between the first estimated magnetic pole position 6g and the first estimated magnetic pole position 6g, and the first estimated magnetic pole position 6e is any of the sum, product, or quotient of the second estimated magnetic pole position 6g. It may be compared with the threshold value θ erth using the calculation result.

The reset instruction in step S202 in the flowchart of FIG. 2 is a reset instruction because the speed of the permanent magnet synchronous motor 300 is a low speed at which the first estimation unit 6a cannot operate stably. The reset instruction in step S205 is a reset instruction for the reason that the step-out of the first estimation unit 6a is detected.

  In FIG. 2, the determination / startup processing unit 6 c may perform only one of the reset instruction in step S <b> 202 at the start time and the reset instruction in step S <b> 205 at the time of step-out, depending on the operating conditions.

  FIG. 3 shows the operation of each part according to the flowchart in FIG. If the absolute value of the second estimated speed 6h is less than the predetermined value ωth, the first estimating unit 6a stops and does not detect step-out. When the absolute value of the estimated speed is equal to or greater than the predetermined value ωth, the first estimating unit 6a is operated to perform step-out detection. It is assumed that the second estimation unit 6b always operates.

  There are the following problems in detecting a failure of a position / speed sensor to which sensorless control is applied. It is known that the observer method is generally difficult to operate stably when the rotational speed of the motor is low. This is because the disturbance voltage such as the inverter dead time voltage becomes relatively higher as the speed becomes lower than the voltage required for driving the permanent magnet synchronous motor. For this reason, it is difficult for the observer to self-activate the observer at the time of activation in which the permanent magnet synchronous motor accelerates from a stopped state to a predetermined speed. In order to stably start the observer when the permanent magnet synchronous motor is started, external motor excitation position information and motor speed information are required. This can be done by using the position / speed sensor information, but the position / speed sensor may have failed before the motor is started. To realize reliable fault detection and stable motor stop control processing when a fault occurs. Needed to start the observer without position velocity sensor information.

  Since the first estimation unit 6a is based on an observer, self-starting from zero speed is difficult. However, if the position / speed sensor 400 is configured to start using the information of the position / speed sensor 400, the position / speed sensor 400 cannot be started if the position / speed sensor 400 is out of order. In the case of the same configuration, if the position / velocity sensor 400 outputs incorrect position information due to a failure, the start is based on the incorrect information, so the first estimated magnetic pole position 6e and the first estimated speed 6f are There is an error, and the position / velocity sensor failure detection and the subsequent stop control process which are the original purposes cannot be performed. Since the second estimation unit 6b does not have a feedback processing structure inside, the second estimation unit 6b can easily start itself, and can output an estimated magnetic pole position and an estimated speed although the accuracy is low at a low speed including zero speed. Since the present invention uses this property to activate the first estimation unit 6a using the estimation information from the second estimation unit 6b, it is possible to eliminate the difficulty of self-starting of the first estimation unit 6a. Moreover, since the position / velocity sensor information is not used when starting the first estimating unit 6a, the original purpose of the position / velocity sensor failure detection and the stop control process can be implemented without any problem. Among permanent magnet synchronous motors, there is a type called an embedded magnet type that uses the change in motor inductance depending on the magnetic pole position and can superimpose a high frequency current on the motor to estimate the magnetic pole position, There is a type called a surface magnet type in which there is no inductance change with respect to the magnetic pole position. The flowchart of FIG. 2 is particularly effective for a surface magnet type permanent magnet synchronous motor. This is because the first estimation unit 6a can be activated even if the magnetic pole position cannot be estimated by superposition of high-frequency current as in the case of an embedded permanent magnet synchronous motor. Needless to say, the flowchart of FIG. 2 can also be applied to an embedded magnet type permanent magnet synchronous motor.

  Next, the effect of performing step-out detection by comparing the first estimated magnetic pole position from the first estimating unit 6a and the second estimated magnetic pole position from the second estimating unit 6b will be described. FIG. 5 shows an example of the first estimated speed, the second estimated speed, the first estimated magnetic pole position, and the second estimated magnetic pole position during motor acceleration operation. The difference between the first estimated speed and the second estimated speed shows an example in which the locus changes depending on the motor acceleration pattern. FIG. 5A and FIG. 5B have different acceleration rates, and FIG. 5B performs a faster acceleration. At this time, based on the difference between the first estimated speed and the second estimated speed, based on FIG. 5 (a), the threshold for detecting the observer step-out of the first estimating unit 6a is set to a speed difference of 300 rpm or more. However, when the acceleration rate changes as described in FIG. 5B, even if no step-out occurs, the above threshold is satisfied and step-out erroneous detection is performed. As shown in FIG. 5, since the trajectory of the difference between the first estimated speed and the second estimated speed is affected by the original motor speed trajectory, in the case where it varies greatly depending on the application destination, it is adjusted to the application destination. Adjustment work becomes enormous. On the other hand, the difference between the first estimated magnetic pole position and the second estimated magnetic pole position remains at most 0.5 [rad] at the time of activation of the first estimation unit, and the waveform is also shown in FIGS. 5 (a) and 5 (b). There is little difference in shape. As described above, this is a result of the high frequency component being suppressed by the integral characteristic between the estimated speed and the estimated magnetic pole position, and step-out detection is performed based on the difference between the first estimated magnetic pole position and the second estimated magnetic pole position. By doing so, it is possible to realize a reliable step-out detection with few false detections. If the observer step-out cannot follow the true motor magnetic pole position, at least the absolute value of the difference between the first estimated magnetic pole position and the second estimated magnetic pole position is 90 degrees (1.5708 [rad]) or more, Since the polarity of each voltage / current signal on the rotation coordinate based on the estimated magnetic pole position is inverted with respect to the true value, it becomes difficult to recover, and this can be defined as step-out. FIG. 6 is an example of signal waveforms of the first estimated magnetic pole position and the second estimated magnetic pole position during observer step-out. At time 0.2 [sec], an offset error is added to the first estimated magnetic pole position of the first estimation unit 6a to simulate a step-out state. After the occurrence of the step-out, the phase of the second estimation unit 6b changes at a predetermined cycle, whereas the phase of the first estimation unit 6a is almost stopped. For this reason, if the difference between the first estimated magnetic pole position and the second estimated magnetic pole position is a sawtooth, and the absolute value thereof is 90 degrees (π / 2 [rad]) or more, the step-out detection threshold value is 2π / ω / 4 [sec]. As described above, it is easy to set a threshold for detecting step-out for comparison of estimated speeds. The threshold setting for the step-out detection described above is an example, and the threshold value for the difference between the estimated magnetic pole positions is increased and the duration threshold value is increased to adjust for suppressing false detection, or vice versa. It is also possible to speed up the key detection time.

  FIG. 4 is a diagram describing the first estimation unit 6a and the second estimation unit 6b in more detail. The first estimation unit 6a includes an observer 6a1, an adaptive identifier 6a2, coordinate converters 6a3 and 6a4, and an integrator 6a5. The second estimation unit 6b includes an induced voltage calculation unit 6b1 and a three-phase / two-phase conversion unit 6b2. , 6b3, arc tangent calculation unit 6b4, and differentiator 6b5.

In the first estimation unit 6a, the input voltage command 12 (Vu ^* , Vv ^* , Vw ^* ) and the detected current 13 (Iu, Iv, Iw) are converted into signals on the rotating coordinates by the coordinate converters 6a3, 6a4. . The voltage command 12 is converted into voltage signals Vd ′ and Vq ′, and the detected current 13 is converted into current signals Id ′ and Iq ′. Coordinate conversion formulas are the following formulas (1) and (2).

  The estimated magnetic pole position 6e of the first estimation unit 6a is used for the coordinate conversion, and is described as θest1 here. The symbol “′” indicates a signal on the estimated magnetic pole position, and is described to distinguish it from the signal on the magnetic pole position of the position / velocity sensor base described in FIG. 1. The observer (also referred to as an observer calculation unit) 6a1 is derived from the circuit equation of the permanent magnet synchronous motor on the rotational coordinates shown in the following expression (3).

Ｌ：インダクタンス、Ｒ：巻線抵抗、ωｒｅ：電気角速度、Φ：誘起電圧定数

L: inductance, R: winding resistance, ωre: electrical angular velocity, Φ: induced voltage constant

  Observer processing is performed by the following formula (4). Various schemes have been developed for the configurations of the observer 6a1 and the adaptive identifier 6a2, and an example is shown here. Estimated currents Idest and Iquest are output with the voltage signals Vd 'and Vq' as inputs. The observer 6a1 also performs a process of feeding back the difference between the estimated currents Idest and Iquest and the current signals Id ′ and Iq ′ by a predetermined multiplication, and the gain h is the estimated response or estimation of the estimated speed of the observer 6a1 or sensorless control. Although it is a parameter for adjusting the estimated response of the magnetic pole position, it is not the essence of the present invention, so the description is omitted. The resetting of the observer internal information indicates a process of overwriting a time-varying estimated variable with a detection value. The first estimated magnetic pole position 6e (θest1) used in the expressions (1) and (2) is overwritten with the second estimated magnetic pole position 6g (θest2). At the same time, the current signals Id ′ and Iq ′ derived from the equation (1) also change. In the following equation (5), at the time of reset, the estimated currents Idest and Iquest are overwritten with the second estimated magnetic pole position 6g (θest2) and the current signals Id ′ and Iq ′ calculated by the equation (1), respectively. The estimated magnetic flux Φdrest is overwritten with the induced voltage constant Φ. The same applies when the observer 6a1 other than the equation (5) is applied, and the time-variable state variable (estimated current) is overwritten with the detected current value or motor constant obtained by coordinate conversion using the second estimated magnetic pole position 6g (θest2). To do.

ｈ１１，ｈ１２，ｈ２１，ｈ２２，ｈ３１，ｈ３２：オブザーバフィードバックゲイン

h11, h12, h21, h22, h31, h32: Observer feedback gain

  The adaptive identifier 6a2 inputs the current estimation errors ΔId and ΔIq and calculates the estimated speed 6f (ωest1). The processing of the adaptive identifier 6a2 is performed by PI (proportional / integral) processing in the integrator 6a5. The following formula (6) is an example. At the time of reset, the component held by the proportional term is overwritten with zero, and the component held by the integral term is overwritten at the estimated speed 6h (ωest2) by the second estimating unit 6b.

Ｋｐ：適応同定器比例ゲイン、Ｋｉ：適応同定器積分ゲイン

Kp: Adaptive identifier proportional gain, Ki: Adaptive identifier integral gain

The configuration of the adaptive identifier 6a2 according to Equation (6) is an example using only the current estimation error ΔIq, and other configuration methods using the current estimation error ΔId are also known, but the portion corresponding to the proportional term is zero. The same operation can be performed by overwriting a place holding past numerical information such as an integrator with the estimated speed 6h (ωest2).
The estimated magnetic pole position 6e (θest1) is obtained by integrating the estimated speed 6f (ωest1) obtained by the equation (6) with the integrator 6a5. In the observer-type sensorless control, a processing loop is formed based on the principle that the current estimation errors ΔId and ΔIq are generated due to an error in the motor speed information included in the observer 6a1, and therefore the estimated speed 6f (ωest1) is the observer. It is also fed back to 6a1. Regarding the observer 6a1 and the adaptive identifier 6a2 for performing the observer processing shown in the equation (4), the configuration and the design of the feedback gain form another technical field, and are not the essence of the present invention. Is omitted. However, as described above, since the mechanism is essentially an estimation mechanism based on feedback, speed estimation and magnetic pole position estimation response are limited, and a problem of step-out occurs.
The 2nd estimation part 6b is based on the circuit equation on the two-phase stationary coordinate of the permanent magnet synchronous motor shown to following formula (7). The induced voltages Eα and Eβ calculated by the induced voltage calculator 6b1 are expressed by the following equation (8).

θｒｅ：電気角（磁極位置）

θre: Electrical angle (magnetic pole position)

  Current signals Iα and Iβ and voltage signals Vα and Vβ for estimating an induced voltage waveform are obtained by the following equations (9) and (10) in the three-phase / two-phase converters 6b2 and 6b3.

  The induced voltage waveform is estimated by the following equation (11) based on the circuit equations shown in equations (7) and (8). The estimated magnetic pole position 6g (θest2) is derived from the estimated induced voltage waveform in the arc tangent calculation unit 6b4 in the following (12). The estimated speed 6h (ωest2) of the second estimating unit 6b can be obtained by time-differentiating the estimated magnetic pole position 6g (θest2) with a differentiator 6b5.

Ｅαｅｓｔ，Ｅβｅｓｔ：推定誘起電圧

Eαest, Eβest: Estimated induced voltage

  The operational effects on the signal processing surface of the operation of the determination / startup processing unit 6c based on the flowchart shown in FIG. 2 are as described below. That is, as a first effect, the out-of-step detection accuracy of the observer of the first estimation unit 6a can be improved by comparing the first estimated magnetic pole position and the second estimated magnetic pole position. Further, when the observer step out is detected, reset processing is performed using the estimated speed and the estimated magnetic pole position information from the second estimation unit 6b of the induced voltage method, so that the recovery from the step out can be performed, and the magnetic pole position and speed are also detected. As the entire operation of the estimator 6, the output of the estimated magnetic pole position and the estimated speed can be continued. Furthermore, by using the estimated speed and the estimated magnetic pole position information from the second estimating unit that can be self-started from when the motor is stopped as the initial value when starting the first estimating unit 6a, it is possible to detect the low speed including when the motor is stopped. The activation of the first estimating unit 6a, which is difficult to operate, can be realized stably and without any other motor position or speed information from the outside.

  The magnetic pole position / speed estimator 6 described above is used in a motor control device that performs position or speed control with high response by control with a position speed sensor such as a drive system for a servo motor. This is particularly effective when fault detection and deceleration stop control are reliably performed. In the servo motor, the position or speed control is performed with a high response of 500 Hz or 1 kHz, so that the motor speed changes greatly and the observer of the first estimation unit 6a easily steps out. The use of the position / velocity estimator 6 can continue to obtain the estimated magnetic pole position and the estimated speed, so that the position / speed sensor 400 can continue to detect a failure. Since the magnetic pole position / speed estimator 6 as a whole does not depend on other external motor position or speed information, the failure detection process can be made robust. In particular, the above-mentioned servo motor often uses a surface magnet type permanent magnet synchronous motor. At this time, unlike the permanent magnet synchronous motor of the embedded magnet type, the magnetic pole position cannot be estimated by superimposing the high-frequency current, but the observer of the first estimating unit 6a can be stably started by using the configuration of the present invention. In deceleration stop control after failure detection, motor control is performed based on the estimated magnetic pole position and estimated speed from the first estimation unit 6a as much as possible. Speed control can be realized.

  As described above, the estimated magnetic pole position has an integral relationship with the estimated speed in both the sensorless control method of the observer method and the induced voltage method. Therefore, due to the motor speed, a transient speed estimation error is included in the difference between the first estimated speed from the first estimator using the observer method and the second estimated speed from the second estimator using the induced voltage method. The high frequency component is suppressed by the integration characteristic in the difference between the first estimated magnetic pole position and the second estimated magnetic pole position. Although this high frequency component has caused a drop in the accuracy of step-out detection, by detecting based on the difference between the first estimated magnetic pole position and the second estimated magnetic pole position, the content of the high frequency component becomes the estimated speed and the estimated magnetic pole. It can be suppressed by the integral characteristic between positions, and the accuracy of step-out detection can be improved.

In addition, the same effect can be obtained for preventing deterioration in accuracy of step-out detection caused by a high-frequency component caused by a high-frequency voltage disturbance included in the second estimated speed.
Furthermore, since the second estimation unit does not have a feedback structure, it is possible to output the estimated magnetic pole position and the estimated speed without requiring external information even when the permanent magnet synchronous motor is started from a stopped state. For this reason, by using the estimation information from the second estimation unit, it is possible to activate the observer of the first estimation unit and realize appropriate failure detection and smooth stop control processing.

  In short, the estimator for estimating the rotor speed and the magnetic pole position of the permanent magnet synchronous motor includes a first estimation unit for obtaining the estimated speed and the estimated magnetic pole position of the permanent magnet synchronous motor by an observer based on the motor model and an adaptive identifier; A second estimating unit for obtaining an estimated speed and an estimated magnetic pole position of the permanent magnet synchronous motor from the motor induced voltage; an estimated magnetic pole position of the first estimating unit and an estimated magnetic pole position of the second estimating unit; or an estimated speed of the second estimating unit When the speed of the permanent magnet synchronous motor is controlled to be highly responsive by providing a determination / startup processing unit that determines whether the estimated speed of the first estimation unit and the estimated magnetic pole position are reset using It is possible to detect the step-out of the observer that estimates the rotor speed and the magnetic pole position with high accuracy.

  In addition, when the determination / activation processing unit detects the step-out of the first estimation unit using the estimated magnetic pole position of the first estimation unit and the estimated magnetic pole position of the second estimation unit, the estimation speed and the estimation of the second estimation unit are detected. By performing the determination of resetting the estimated speed of the first estimating unit and the estimated magnetic pole position using the magnetic pole position, when the speed of the permanent magnet synchronous motor is controlled to a high response, an observer step-out occurs. In addition, the estimation of the magnetic pole position can be continued for the rotor speed of the motor.

  In addition, the determination / activation processing unit uses the estimated speed and the estimated magnetic pole position of the first estimating unit when the absolute value of the estimated speed of the second estimating unit exceeds a predetermined value. In the control system for the permanent magnet synchronous motor having no magnetic saliency, the position speed sensor connected to the permanent magnet synchronous motor (by determining the resetting of the estimated speed and the estimated magnetic pole position and starting the first estimating unit) The observer can be activated without using the information of the position detector.

  In the first embodiment, when the determination / activation processing unit detects a step-out of the first estimation unit using the estimated magnetic pole position of the first estimation unit and the estimated magnetic pole position of the second estimation unit, (2) When the estimated speed and the estimated magnetic pole position of the first estimating unit are reset using the estimated speed and the estimated magnetic pole position of the estimating unit, and the absolute value of the estimated speed of the second estimating unit exceeds a predetermined value In the above, the configuration for performing both the determination of resetting the estimated speed and the estimated magnetic pole position of the first estimating unit using the estimated speed and the estimated magnetic pole position of the second estimating unit is described. Since the detection function and the activation function of the first estimation unit are provided, the functions are independent of each other and may be used by only one of them.

  Needless to say, the configuration of the present invention described above can be applied to a wound field type synchronous motor regardless of the model of the permanent magnet synchronous motor.

Embodiment 2. FIG.
A second embodiment of the present invention will be described. The configuration of the present invention is particularly effective under the conditions described below for the current controller 2 in the motor control device. The inverter 3 operates in response to the switching command 14 from the PWM processing unit 2d in the current controller 2 shown in FIG. 1, and as a result, a rectangular wave voltage is applied to the permanent magnet synchronous motor 300. This rectangular wave voltage includes a PWM ripple component in addition to a component for driving the permanent magnet synchronous motor. As a result, the current flowing through the motor includes ripples caused by PWM. The switching command 14 is generated by comparing the voltage command 12 with the carrier wave. As a carrier wave, a triangular wave is often selected. When the current controller 2 takes in the detected current (signal) 13 from the current sensor 4, sampling is performed at a predetermined cycle, but in order to avoid the current ripple being included in the current sampling value, a triangular wave Current sampling is performed in synchronization with peak timing or valley timing. However, in some cases, the operation of the current controller 2 may not be synchronized with the peak / valley timing of the triangular wave, or may operate at a higher speed than the triangular wave. At this time, the detection current 13 includes many current ripples. Further, since the current controller 2a of the current controller 2 operates to cancel this current ripple, the voltage command 12 also pulsates. Therefore, a large amount of disturbance signals due to PWM are also input to the magnetic pole position / speed estimator 6 via the voltage command 12 and the detection current 13. Here, in the process of estimating the magnetic pole position in the second estimation unit 6b, differentiation of the current is included as is apparent from the equation (11). Further, since the second estimated speed ωest2 is also derived by the differentiation of the second estimated magnetic pole position θest2, the second estimated speed ωest2 includes many disturbance components. FIG. 7 shows an example in which the second estimated speed pulsates due to current ripple caused by PWM. FIG. 7 shows an example of the operation, which is the result when the carrier frequency is 4.5 kHz and the operation cycle of the current controller 2 is 100 kHz. The current controller 2 operates at a period earlier than that of the triangular wave and is not synchronized with the peaks and valleys. Therefore, according to the above process, the second estimated speed ωest2 pulsates despite the motor speed being constant, and the first estimated speed ωest1 and the second estimated speed ωest2 are also pulsating. There is a possibility of false detection. However, the difference between the first estimated magnetic pole position θest1 and the second estimated magnetic pole position θest2 is maintained at 0.5 [rad] or less, and step-out detection is performed based on the difference between the magnetic pole positions. If such a step-out detection threshold is used, erroneous detection can be prevented.

  In the drive system for servo motors, for the purpose of increasing the response of the position control response or the response of the speed control response, the response control of the current control, which is a minor loop thereof, is increased. At this time, the current controller 2 operates at a period higher than that of the triangular wave as the carrier, and the operation state as shown in FIG. Even in such a case, it is possible to suppress erroneous detection of step-out by comparing the first estimated magnetic pole position θest1 and the second estimated magnetic pole position θest2.

Embodiment 3 FIG.
Next, a third embodiment of the present invention will be described. FIG. 8 shows the configuration of the motor control apparatus according to the third embodiment.
In the first embodiment, each component having a specific function in the motor control device described in FIG. 1 operates in cooperation with each other, and solves the problem. However, the program executed on the processor As a result of the signal processing and the signal processing in the logic circuit provided on the processor, a function may be realized. FIG. 8 is a configuration diagram thereof, and the processor 23 in the figure reads the program from the storage device 24 and executes it. The processor 23 writes or reads information to be temporarily stored in the course of the processing.

The processing realized by executing the program by the processor 23 is the speed controller 1, the current controller 2, the inverter 3, the current sensor 4, the speed converter 7, and the magnetic pole position converter 8 shown in FIG. This is the same as the signal switch 5, the failure detector 9, and the magnetic pole position / speed estimator 6. Estimated magnetic pole position 20 (θest), estimated speed 21 (ωest), coordinate conversion phase 16 (θ), feedback speed 15 (ω), failure detection signal 22, current command 11, detected magnetic pole position 19 (θdet), detected speed 18 (ωdet), the voltage command 12 is realized as information temporarily stored in the storage device 24. The speed command 10 is transmitted from the host controller 200 by communication, but is similarly realized as information temporarily stored in the storage device 24. Since the PWM processing unit 2d in the current controller 2 is realized by a logic circuit provided on the processor, the switching command 14 is a signal on the logic circuit. The failure detection signal 22 is transmitted from the processor 23 to the host controller 200 by communication. As a result of signal processing of a program executed on the processor and signal processing in a logic circuit provided on the processor,
Even in the configuration in which the function is realized, the effect as shown in the first embodiment can be obtained.

  The present invention can be freely combined with each other within the scope of the present invention, or can be appropriately modified or omitted.

DESCRIPTION OF SYMBOLS 1 Speed controller, 2 Current controller, 6 Magnetic pole position and speed estimation part, 6a 1st estimation part, 6a1 observer, 6a2 Adaptive identification part, 6b 2nd estimation part, 6c Judgment / start-up process part, 9 Fault detector, 100 Motor control device

Claims (3)

  A motor controller having an estimator for estimating a rotor speed and a magnetic pole position of a permanent magnet synchronous motor, wherein the estimator is estimated and estimated by an observer based on a motor model and an adaptive identifier. A first estimating unit for obtaining a magnetic pole position, a second estimating unit for obtaining an estimated speed and an estimated magnetic pole position of a permanent magnet synchronous motor from a motor induced voltage, an estimated magnetic pole position of the first estimating unit, and the second estimating unit And a determination / startup processing unit for determining whether to reset the estimated speed of the first estimating unit and the estimated magnetic pole position using at least one of the estimated magnetic pole position and the estimated speed of the second estimating unit. A motor control device.   When the determination / startup processing unit detects a step-out of the first estimation unit using the estimated magnetic pole position of the first estimation unit and the estimated magnetic pole position of the second estimation unit, the second estimation unit estimates The motor control device according to claim 1, wherein a determination is made to reset the estimated speed and the estimated magnetic pole position of the first estimating unit using the speed and the estimated magnetic pole position. The determination / activation processing unit uses the estimated speed and the estimated magnetic pole position of the second estimating unit when the absolute value of the estimated speed of the second estimating unit exceeds a predetermined value. 2. The motor control device according to claim 1, wherein a determination is made to reset the estimated speed and the estimated magnetic pole position of the unit, and the first estimating unit is activated.

Images