JP5891986B2 - Motor control device - Google Patents

Description

  The present invention relates to a motor control device. More specifically, the present invention relates to a motor that can ensure the frequency of origin learning even when trying to improve the learning accuracy by executing the origin learning of the resolver in a state where various parameters fall within a predetermined range. The present invention relates to a control device.

  In response to the recent increase in awareness regarding energy saving and global environment protection, for example, electric vehicles equipped with a motor as a power source are spreading. An example of such a motor is a three-phase AC permanent magnet type synchronous motor. Further, such a motor is driven by a rotating magnetic field corresponding to the position (rotation angle) of the rotor. Therefore, in order to correctly control the rotation of the motor, it is important to correctly detect the rotation angle of the rotor of the motor. Examples of the rotation angle detection means for detecting the rotation angle of the rotor of the motor include a resolver and an encoder.

  Note that a resolver is generally used as a rotation angle detection means provided in a motor used as a power source of an electric vehicle such as an electric vehicle (EV), a hybrid vehicle (HV), and a plug-in hybrid vehicle (PHV). Yes. Generally, the resolver includes a rotor that rotates in conjunction with a rotation shaft of a motor, and is configured to detect the position of the rotor of the motor by the rotor.

  However, in reality, a difference (offset) is generated between the position of the rotor of the motor detected by the resolver and the actual position of the rotor of the motor due to various causes such as a shift of the resolver mounting position. There is a case. In such a state, it is difficult to correctly control the rotation of the motor in a state where the offset remains generated. Therefore, in this technical field, a motor that detects the offset of the resolver and is detected by the resolver according to the detected offset. Various attempts have been proposed to correct the rotor position.

  The detection of the offset of the resolver and the correction by the detected offset (also referred to as “origin learning”) can be performed, for example, in a state where the motor is rotating with a torque of 0 (zero). When there is no resolver offset in such a state, both the excitation current Id and the torque current Iq in the dq axis coordinates are 0 (zero). Accordingly, in a state where the torque is 0 (zero) and the motor is rotating, the excitation voltage Vd = 0 (zero) and the torque are obtained from the voltage equation shown in the following equation (1) as shown in the equation (2). The voltage Vq = ωφ is derived.

  Here, the above will be further described with reference to the accompanying drawings. FIG. 1 is a schematic diagram for explaining the difference between the excitation voltage Vd and the torque voltage Vq detected when the torque is 0 (zero) and the resolver is offset in the state where the motor is rotating. When the state represented by Expression (2) is expressed in dq axis coordinates, it is as shown in FIG. However, when there is a resolver offset, the control recognition axis shifts as represented by the broken line shown in FIG. 1B, and the original excitation voltage Vd (d-axis), which is 0 (zero) in the original case. The component Vd ') is detected. In the resolver origin learning, the resolver offset is detected based on the excitation voltage Vd thus detected.

  For example, in this technical field, when a phase correction command is input, a torque command is ignored and a d-axis current command and a q-axis current command are set to zero, and a d-axis-q-axis current command A current controller that outputs a d-axis-q-axis voltage command based on the phase correction amount detector that obtains an offset amount to zero when the d-axis voltage command is not zero when the phase correction command is input; A synchronous motor control device comprising: an adder for adding the rotor position angle and the offset amount; and a voltage converter for obtaining a three-phase voltage command based on the added value and the d-axis-q-axis voltage command. It has been proposed to accurately correct a rotational position shift related to a rotational position detector even when a load is connected to the rotor without requiring a load detector or a motor voltage detector (for example, patents). Reference 1 See beauty 2).

  Also, in this technical field, paying attention to the fact that the waveform of the induced voltage of the motor always crosses zero when the actual electrical angle of the motor is 0 degree, regardless of the presence / absence of the external load applied to the motor. By obtaining the actual detected value θ1 of the resolver when the waveform of the induced voltage output from the motor zero-crosses as the offset value θ2, the offset value θ2 corresponding to the assembly error generated between the resolver and the motor is more accurately obtained than in the past. It is also proposed to specify (see, for example, Patent Document 3).

  Further, the present inventor has found that various parameters (for example, the motor rotation speed, the system voltage VH that is an input voltage to the inverter, the inverter carrier), which are factors that affect the accuracy of the resolver origin learning in the motor control device. Frequency), a range suitable for improving the origin learning accuracy of the resolver is determined, and the resolver origin learning is executed only when the parameter to be controlled is within the range. It has been separately found that the accuracy of origin learning can be improved (a separate application is filed by the present inventor).

  However, if the range of various parameters is limited as a condition for executing the resolver origin learning as described above, all conditions for executing the resolver origin learning may fall within a suitable range to execute the origin learning. There is a possibility that the frequency of performing the origin correction of the motor rotor may be reduced, or the origin correction of the motor rotor may not be performed.

  That is, in this technical field, the motor control device ensures the frequency of origin learning even when trying to improve the learning accuracy by executing the origin learning of the resolver in a state where various parameters fall within a predetermined range. There is a need for technology that can do this.

JP 2004-129359 A JP 2004-266935 A JP 2006-296025 A

  As described above, in the technical field, even when the motor control device attempts to improve the learning accuracy by executing the origin learning of the resolver in a state where various parameters fall within a predetermined range, There is a need for technology that can ensure the frequency. The present invention has been made to meet such a demand. That is, the present invention secures the frequency of origin learning even when the motor control device attempts to improve learning accuracy by executing resolver origin learning in a state where various parameters fall within a predetermined range. Is one purpose.

The above purpose is
Based on the rotation angle of the rotor of the motor detected by the resolver, the rotation of the motor is controlled by controlling the voltage value and phase of the command voltage from the inverter including the switching element to the motor,
When the motor is rotating and the required torque for the motor is 0 (zero), the origin of the resolver is learned,
The resolver origin learning is performed only in a state where the rotational speed of the motor is within a predetermined range when a dead time is provided in which the electrical continuity between the upper arm and the lower arm of the inverter is simultaneously cut off. Execute learning permission judgment to allow
In the motor control device,
If the resolver origin learning is allowed,
When the rotational speed of the motor falls within a first range defined by the rotational speed of the motor, the origin learning of the resolver is performed,
When the rotational speed of the motor does not fall within the first range and falls within a second range defined by the rotational speed of the motor that extends over a range wider than the first range, the origin learning of the resolver is performed. And repeating the process after the learning permission determination,
If the rotational speed of the motor does not fall within the second range, the origin of the resolver is not learned.
This is achieved by the motor controller.

  According to the present invention, in the motor control device, the frequency of origin learning is ensured even when trying to improve learning accuracy by executing resolver origin learning in a state where various parameters fall within a predetermined range. Can do.

It is a schematic diagram explaining the difference by the presence or absence of the offset of the excitation voltage Vd and torque voltage Vq which are detected in the state which the torque is 0 (zero) and the motor is rotating. It is a schematic diagram explaining the difference of the relationship between the switching instruction | command by the direction of the electric current in a dead time, and the output of an inverter. It is a schematic diagram explaining the influence of the system voltage VH which is the input voltage to an inverter with respect to the fall of the precision of the origin learning of a resolver resulting from providing dead time. It is a typical graph explaining the high precision area | region and low precision area | region prescribed | regulated by the combination of the system voltage VH and the rotation speed of a motor in the motor control apparatus which concerns on one embodiment of this invention. It is a schematic diagram explaining the influence of the carrier frequency of an inverter with respect to the fall of the precision of the resolver origin learning resulting from providing a dead time. It is a flowchart showing the flow of the various processes in the origin learning of the resolver performed in the motor control apparatus which concerns on one embodiment of this invention.

  As described above, one object of the present invention is to provide a motor control apparatus that can improve the learning accuracy even when attempting to improve the learning accuracy by executing the resolver origin learning in a state where various parameters fall within a predetermined range. It is to ensure the frequency of learning. As a result of earnest research to achieve the above object, the present inventor performs origin learning to increase the learning frequency even when the value of the parameter falls within a range where the learning accuracy is lower than the desired learning accuracy, In such a case, it is possible to secure the frequency of origin learning while improving the accuracy of origin learning of the resolver by performing origin learning again and performing control so that origin learning is performed with a desired learning accuracy. I found.

  More specifically, the present inventor defines a first range suitable for achieving a desired learning accuracy for a predetermined parameter and a second range wider than the first range, and the parameter is the first range. If the parameter is within the first range, the resolver origin learning is performed. If the parameter is not within the first range and the second range is entered, the resolver origin learning is performed and the origin learning is performed again. When it does not fall within the second range, it has been found that the origin learning frequency can be secured while improving the origin learning accuracy of the resolver by not performing the origin learning, and the present invention has been conceived. is there.

That is, the first embodiment of the present invention is:
Based on the rotation angle of the rotor of the motor detected by the resolver, the rotation of the motor is controlled by controlling the voltage value and phase of the command voltage from the inverter including the switching element to the motor,
When the motor is rotating and the required torque for the motor is 0 (zero), the origin of the resolver is learned,
The resolver origin learning is performed only in a state where the rotational speed of the motor is within a predetermined range when a dead time is provided in which the electrical continuity between the upper arm and the lower arm of the inverter is simultaneously cut off. Execute learning permission judgment to allow
In the motor control device,
If the resolver origin learning is allowed,
When the rotational speed of the motor falls within a first range defined by the rotational speed of the motor, the origin learning of the resolver is performed,
When the rotational speed of the motor does not fall within the first range and falls within a second range defined by the rotational speed of the motor that extends over a range wider than the first range, the origin learning of the resolver is performed. And repeating the process after the learning permission determination,
If the rotational speed of the motor does not fall within the second range, the origin of the resolver is not learned.
It is a motor control device.

  As described above, the motor control device according to the present embodiment controls the voltage value and phase of the voltage command from the inverter including the switching element to the motor based on the rotation angle of the rotor of the motor detected by the resolver. Thus, the rotation of the motor is controlled. The configuration of the motor is not particularly limited, and as a specific example, for example, it is widely used as a power source for electric vehicles such as electric vehicles (EV), hybrid vehicles (HV), plug-in hybrid vehicles (PHV), and the like. And three-phase AC permanent magnet type synchronous motor. As described above, since such a motor is driven by a rotating magnetic field corresponding to the position (rotation angle) of the rotor, in order to properly control the rotation of the motor, the rotation angle of the rotor of the motor is set correctly. It is important to detect.

  Specific examples of the rotation angle detection means for detecting the rotation angle of the rotor of the motor include, for example, a resolver and an encoder as described above. In the motor control device according to this embodiment, A resolver widely used as a rotation angle detection means provided in a motor used as a power source of a vehicle is employed as the rotation angle detection means. As described above, the resolver generally includes a rotor that rotates in conjunction with the rotation shaft of the motor, and is configured to detect the position of the rotor of the motor by the rotor.

  By the way, as described above, in the motor control apparatus, even when trying to improve the learning accuracy by executing the origin learning of the resolver in a state where various parameters fall within a predetermined range, One purpose is to ensure the frequency. Here, as a factor of lowering the learning accuracy in the origin control of the resolver in the motor control device including the inverter, for example, the setting of the dead time in the inverter can be mentioned.

  As is well known to those skilled in the art, the dead time is a control signal for switching control of two switching elements (upper arm and lower arm) constituting an inverter chopper circuit that supplies a voltage applied to each phase of the motor. In FIG. 2, the period is set for the purpose of preventing the upper arm and the lower arm from being simultaneously turned on (ON), and is a period in which the upper arm and the lower arm are simultaneously turned off (OFF). By providing such a dead time, it is possible to prevent the upper arm and the lower arm from being simultaneously turned on (ON), causing an excessive short-through current to flow and causing problems such as damage to the power supply circuit and the switching element, for example. it can.

  However, while having the above-mentioned advantages, by providing dead time, the duty ratio of the upper arm and the lower arm can be calculated based on the command duty ratio calculated based on the target output voltage, and the upper arm and lower arm can actually be There is a difference between the duty ratio (actual duty ratio) when the arm is turned on / off. This point will be described below in more detail with reference to FIG.

  FIG. 2 is a schematic diagram for explaining the difference in the relationship between the switching command (hereinafter sometimes referred to as “SW command”) depending on the direction of the current in the dead time and the output of the inverter, as described above. First, as shown in FIG. 2A, when the direction of the current is positive (that is, the direction of flow from the inverter to the phase coil of the motor), the lower arm is turned off (OFF) with respect to the rise of the SW command. ) Although the command is not delayed, the upper arm on (ON) command is delayed by the dead time. As a result, the rise of the actual voltage is delayed by the dead time with respect to the rise of the SW command. On the other hand, although the lower arm on command is delayed by the dead time, the upper arm off command is not delayed with respect to the fall of the SW command. As a result, the fall of the actual voltage is not delayed with respect to the fall of the SW command. Thus, when the direction of the current is positive, the actual voltage rise is delayed and the fall is not delayed with respect to the SW command. As a result, when the current direction is positive, the actual duty ratio becomes smaller than the command duty ratio.

  Also, as shown in FIG. 2B, when the current direction is negative (that is, the direction of flow from the motor phase coil to the inverter), the upper arm is turned on (ON) with respect to the rise of the SW command. ) The command is delayed by the dead time, but the lower arm OFF command is not delayed. As a result, the rise of the actual voltage is not delayed with respect to the rise of the SW command. On the other hand, although the upper arm off command is not delayed with respect to the fall of the SW command, the lower arm on command is delayed by the dead time. As a result, the fall of the actual voltage is delayed by the dead time with respect to the fall of the SW command. Thus, when the current direction is negative, the actual voltage rise is not delayed and the fall is delayed with respect to the SW command. As a result, when the current direction is negative, the actual duty ratio becomes larger than the command duty ratio.

  Further, as shown in FIG. 2C, when the magnitude of the current is approximately 0 (zero) and the direction of the current cannot be specified as described above, the dead time at the rise and fall of the SW command In FIG. 5, the influence of the induced voltage corresponding to the electrical angle of the rotor appears in the actual voltage. Specifically, with respect to the rise of the SW command, the upper arm on (ON) command is delayed by the dead time, and the lower arm off (OFF) command is not delayed. As a result, since the magnitude of the current is approximately 0 (zero), the actual voltage should not rise in the dead time. However, due to the influence of the induced voltage corresponding to the electrical angle of the rotor. The actual voltage will increase. However, during the dead time, as described above, the upper arm ON command has not been issued, so the voltage from the upper arm is not applied, and the actual voltage equivalent to the SW command is not applied. Therefore, in the dead time, the actual duty ratio becomes smaller than the command duty ratio by the difference between the voltage to be applied from the upper arm and the induced voltage.

  On the other hand, with respect to the fall of the SW command, the upper arm off command is not delayed, and the lower arm on command is delayed by the dead time. As a result, since the magnitude of the current is approximately 0 (zero), the actual voltage should normally be maintained during the dead time, but the influence of the induced voltage corresponding to the electrical angle of the rotor. As a result, the actual voltage decreases. As a result, therefore, in the dead time, the actual duty ratio becomes larger than the command duty ratio by the amount of the actual voltage that remains without being lowered due to the influence of the induced voltage.

  As described above, since the difference between the command duty ratio and the actual duty ratio differs depending on the direction of the current flowing between each phase of the motor and the corresponding chopper circuit, the actual duty ratio is changed when the current flowing direction is reversed. It will fluctuate discontinuously. As a result, as the dead time included in the period during which the resolver origin learning is executed increases, the accuracy of the resolver origin learning decreases due to the fluctuation of the actual duty ratio as described above.

  The resolver origin learning in the motor control device is executed in a state where the motor is rotating and the required torque for the motor is 0 (zero). At this time, in the motor control device according to the related art, the origin learning of the resolver is executed at the rotation speed of the motor in a state where the required torque for the motor is 0 (zero). That is, in the motor control device according to the prior art, the number of revolutions of the motor when the origin learning of the resolver is executed becomes small, and the accuracy of the origin learning of the resolver is affected by the number of revolutions of the motor. there were.

  Therefore, as a result of studying the relationship between the rotational speed of the motor when the resolver origin learning is performed and the accuracy of the resolver origin learning, the inventor conducted the origin learning of the resolver as described above. By allowing the resolver origin learning only when the motor speed is within a predetermined range, the influence of dead time on the resolver origin learning accuracy can be reduced and the origin learning accuracy can be improved. I found out that I can do it.

  Specifically, if the rotational speed of the motor is too low, the torque voltage Vq in the dq coordinate as described above with reference to FIG. 1 is small, and the sensitivity to the offset of the resolver is low (the excitation voltage is not large unless the offset is very large). It is difficult to detect this as a change in Vd). On the other hand, if the rotational speed of the motor is too high, the ratio of dead time to one rotation period becomes high, and as a result, the accuracy of the learning of the resolver origin caused by providing the dead time becomes remarkable, which is desirable. Absent. Thus, in the inverter that supplies the voltage applied to each phase of the motor, in order to perform the origin learning of the resolver with high accuracy even when the dead time is provided, the rotation speed of the motor is within a predetermined range. It is desirable to be in

  Therefore, in the motor control device according to the present embodiment, the upper arm of the inverter is used to perform the origin learning of the resolver when the motor is rotating and the required torque for the motor is 0 (zero). When the dead time, which is a period in which the electrical continuity of the lower arm and the lower arm are cut off at the same time, is provided, the origin learning of the resolver is permitted only when the rotational speed of the motor is within a predetermined range. Thereby, according to the motor control device according to the present embodiment, even when the dead time is provided in the inverter that supplies the voltage applied to each phase of the motor, the accuracy of the origin learning of the resolver in the motor control device is improved. Can be made.

  Incidentally, the lower limit of the predetermined range for the rotational speed of the motor is, for example, as described above, the torque voltage Vq in the dq coordinate becomes small, and the sensitivity to the resolver offset becomes low. It can be determined as a minimum value in a category in which the decrease in the accuracy of the origin learning of the resolver accompanying the difficulty in being detected as a change in the voltage Vd is within an allowable range. On the other hand, the upper limit of the predetermined range for the number of revolutions of the motor is, for example, as described above, the decrease in the accuracy of the resolver origin learning accompanying an increase in the proportion of dead time per rotation period. It can be determined as the maximum value in a category that falls within the allowable range.

  However, as described above, if the range of the rotational speed of the motor is limited as a condition for executing the origin learning of the resolver as described above, the probability that the origin learning can be performed is reduced, and the origin correction of the motor rotor is performed. There is a possibility that the frequency of performing the operation will be reduced, or that the origin of the rotor of the motor cannot be corrected.

Therefore, in the motor control device according to this embodiment, as described above,
If the resolver origin learning is allowed,
When the rotational speed of the motor falls within a first range defined by the rotational speed of the motor, the origin learning of the resolver is performed,
When the rotational speed of the motor does not fall within the first range and falls within a second range defined by the rotational speed of the motor that extends over a range wider than the first range, the origin learning of the resolver is performed. And repeating the process after the learning permission determination,
If the rotational speed of the motor does not fall within the second range, the resolver origin learning is not performed.

  For example, in the motor control device according to the present embodiment, the parameter corresponds to a range (first range) of parameters (motor rotational speed) that can execute resolver origin learning with desired learning accuracy. In this case, the origin learning is executed. In addition, even when the parameter does not fall within the range (first range), the parameter (in the second range) has a lower learning accuracy than the range (first range). The origin learning is also performed when the motor rotation number) is applicable. Thereby, the frequency of origin learning can be increased. However, this alone reduces the accuracy of origin learning. Therefore, in the motor control device according to the present embodiment, when the resolver origin learning is executed when the parameter does not fall within the first range but falls within the second range, the accuracy becomes high again. Retry origin learning to execute origin learning. Thereby, the fall of the learning precision accompanying the increase in learning frequency can be suppressed. If the parameter does not even fall within the second range, the origin learning is not executed.

  As described above, in the motor control device according to the present embodiment, even when trying to improve the learning accuracy by executing the origin learning of the resolver in a state where the parameter (the number of rotations of the motor) falls within a predetermined range. The frequency of origin learning can be ensured. As a matter of course, the second range defined by the rotational speed of the motor is preliminarily set in advance with respect to the rotational speed of the motor that is a determination criterion for determining whether or not to permit the origin learning of the resolver in the above-described learning permission determination. It is desirable to be equal to or narrower than the determined range.

  In addition, the first range, the second range, and the like of the parameters serving as the determination criteria for determining whether or not to perform the origin learning of the resolver in the motor control device according to the present embodiment are, for example, storage devices (for example, , HDD (Hard Disk Drive), RAM (Random Access Memory), ROM (Read Only Memory), etc.) are stored as a data map or data table, and the origin of the resolver is permitted in the motor controller. It can be configured to be referred to when the determination is made. Alternatively, the first range, the second range, and the like may be defined as functions, for example.

  Incidentally, the accuracy of the resolver origin learning in the motor control apparatus is influenced not only by the motor rotation speed described above but also by various other factors. Therefore, in order to perform the origin learning of the resolver with higher accuracy, it is desirable to appropriately control various factors that affect the learning accuracy and perform the origin learning in a situation where higher learning accuracy can be obtained. . As an example of such a factor, for example, a system voltage VH that is an input voltage to the inverter can be cited. In this case, by performing control so that the system voltage VH falls within an appropriate range, it is possible to perform the origin learning of the resolver with higher accuracy.

  Specifically, if the system voltage VH, which is the input voltage to the inverter, is too low, the torque generated by the motor is reduced, the motor rotation becomes unstable, and the motor rotation speed cannot be kept constant. There is a fear. Thus, if the motor rotation becomes unstable, the accuracy of the resolver origin learning also decreases. Therefore, when performing the resolver origin learning, a voltage higher than the voltage necessary to maintain the motor rotation speed is required. It is desirable to set the value as a target voltage value.

  On the other hand, as disclosed in another application by the present inventor, the higher the system voltage VH that is the input voltage to the inverter is, the more noticeably the resolver's origin learning accuracy decreases due to the provision of dead time. The present inventors have found that

  Here, the influence of the system voltage VH, which is the input voltage to the inverter, on the decrease in the accuracy of the resolver origin learning caused by providing the dead time will be described in detail below with reference to the accompanying drawings. FIG. 3 is a schematic diagram for explaining the influence of the system voltage VH, which is the input voltage to the inverter, on the decrease in the accuracy of the resolver origin learning caused by providing the dead time as described above. As described above with reference to FIG. 2, in the duty ratio of the upper arm and the lower arm of the inverter, by providing a dead time, the command duty ratio calculated based on the target output voltage, and the actual upper arm and There is a difference between the duty ratio (actual duty ratio) when the lower arm is turned on / off.

  In the above, as shown in FIG. 3 (1), when the system voltage VH, which is the input voltage to the inverter, is low, the difference generated between the command duty ratio and the actual duty ratio by providing the dead time (FIG. 3). The shaded area in is small. On the other hand, as shown in FIG. 3 (2), when the system voltage VH, which is the input voltage to the inverter, is high, the difference between the command duty ratio and the actual duty ratio by providing the dead time (in FIG. 3) The shaded area is large. As described above, the higher the system voltage VH that is the input voltage to the inverter, the more remarkable the decrease in the accuracy of the resolver origin learning caused by providing the dead time.

  Therefore, in order to perform the origin learning of the resolver with higher accuracy, the system voltage VH, which is the input voltage to the inverter, is set to the minimum required to achieve the rotational speed of the motor when performing the origin learning. In addition to controlling the voltage to be equal to or higher than the voltage, the accuracy of resolver origin learning is unacceptable in consideration of the decrease in resolver origin accuracy caused by providing the dead time as described above. It is desirable to control so as not to reach a voltage that decreases to an extent.

  In this case, the system voltage VH and the number of rotations of the motor correspond to a range (first range) (high accuracy region) in which a desired learning accuracy is obtained, which is defined by a combination of the system voltage VH and the number of rotations of the motor. By executing the resolver origin learning only in this case, the resolver origin learning can be performed with higher accuracy. However, as described above, if the number of parameters used as the condition for executing the resolver origin learning is increased, the probability that all the parameters fall within a suitable range for executing the resolver origin learning decreases. As a result, the frequency at which the resolver origin learning is performed decreases, and the frequency at which the motor rotor origin correction is performed may decrease, or the motor rotor origin correction may not be performed.

  Accordingly, even in such a case, as described above, the first range (high accuracy region) suitable for achieving the desired learning accuracy for these parameters (system voltage VH and motor rotation speed) and the first If the second range (low accuracy region) wider than the range is defined, and these parameters do not correspond to the high accuracy region but fall within the low accuracy region, the origin learning of the resolver is performed and the origin learning is performed. It is desirable to suppress the decrease in the accuracy of the resolver origin learning while ensuring the frequency of the resolver origin learning by executing again.

That is, the second embodiment of the present invention is:
A motor control device according to the first embodiment of the present invention,
If the resolver origin learning is allowed,
The system voltage VH is manipulated by changing the target voltage value of the system voltage VH, which is the input voltage to the inverter,
When the system voltage VH and the rotation speed of the motor fall within a first range defined by a combination of the system voltage VH and the rotation speed of the motor, the origin learning of the resolver is performed,
The system voltage VH and the rotation speed of the motor are defined by a combination of the system voltage VH and the rotation speed of the motor that do not fall within the first range and extend over a wider range than the first range. When entering the range of, the origin of the resolver is learned, and the process after the learning permission determination is repeated,
When the system voltage VH and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.
It is a motor control device.

  As described above, in the motor control device according to the present embodiment, when the origin learning of the resolver is permitted, the target voltage value of the system voltage VH that is the input voltage to the inverter is changed to change the system voltage. Operate VH. Thus, the origin of the resolver is learned by controlling the system voltage VH to be the target voltage value. As a result, in the motor control device according to this embodiment, the origin of the resolver is executed with higher accuracy by controlling the system voltage VH, which is the input voltage to the inverter, to be an appropriate voltage value. Can do.

  However, as described above, if the range of the system voltage VH and the rotation speed of the motor is limited as a condition for executing the origin learning of the resolver as described above, the probability that the origin learning can be performed becomes low, and the rotation of the motor There is a possibility that the frequency of correcting the origin of the child may be reduced, or the origin of the rotor of the motor cannot be corrected.

Therefore, in the motor control device according to this embodiment, as described above,
If the resolver origin learning is allowed,
When the system voltage VH and the rotation speed of the motor fall within a first range defined by a combination of the system voltage VH and the rotation speed of the motor, the origin learning of the resolver is performed,
The system voltage VH and the rotation speed of the motor are defined by a combination of the system voltage VH and the rotation speed of the motor that do not fall within the first range and extend over a wider range than the first range. When entering the range of, the origin of the resolver is learned, and the process after the learning permission determination is repeated,
When the system voltage VH and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.

  As described above, in the motor control device according to the present embodiment, the determination as to whether or not to perform the resolver origin learning is performed based on the system voltage VH that is the input voltage to the inverter and the rotational speed of the motor. More specifically, in the motor control device according to the present embodiment, the range (first voltage of the system voltage VH and the number of rotations of the motor) that can execute the origin learning of the resolver with a desired learning accuracy (first voltage). If these parameters fall under (range), the origin learning is executed. In addition, even when these parameters do not fall within the above range (first range), the learning accuracy is low compared to the above range (first range), but these are within a wide range (second range). The origin learning is also performed when the parameters (system voltage VH and motor rotation speed) are applicable. Thereby, the frequency of origin learning can be increased.

  However, the accuracy of the origin learning is lowered only by controlling as described above. Therefore, in the motor control device according to the present embodiment, when the resolver origin learning is executed in a case where these parameters do not fall within the first range but fall within the second range, high accuracy is again obtained. Retry origin learning to perform origin learning at. Thereby, the fall of the learning precision accompanying the increase in learning frequency can be suppressed. If these parameters do not even fall within the second range, origin learning is not executed. Thereby, in the motor control device according to the present embodiment, even when trying to improve the learning accuracy by executing the origin learning of the resolver in a state where the system voltage VH and the rotational speed of the motor are in a predetermined range, The frequency of origin learning can be secured.

  In the first range and the second range defined by the combination of the system voltage VH and the motor speed, the upper limit value and / or lower limit value of any one parameter is a parameter other than the one parameter. It may be determined independently of the value of. Alternatively, in the first range and the second range defined by the combination of the system voltage VH and the rotational speed of the motor, the upper limit value and / or the lower limit value of any one parameter is a parameter other than the one parameter. It may vary depending on the value of.

  Here, the first range and the second range defined by the combination of the system voltage VH and the rotation speed of the motor will be described in detail below with reference to the accompanying drawings. FIG. 4 is a schematic diagram for explaining the high accuracy region and the low accuracy region defined by the combination of the system voltage VH and the motor rotation speed in the motor control device according to one embodiment of the present invention as described above. It is a graph. As shown in FIG. 4 (a), in the motor control device according to this embodiment, the high-accuracy region (first range) suitable for performing the origin learning of the resolver with a desired accuracy is shaded in dark. It is represented. In the combination of the system voltage VH corresponding to such a high accuracy region and the rotational speed of the motor, it is possible to perform the origin learning of the resolver with a desired accuracy. Therefore, if only the improvement of the learning accuracy is considered, the system voltage VH If the origin learning is executed only when the rotational speed of the motor falls in the high accuracy region, the accuracy of the resolver origin learning can be improved.

  However, as described above, if the system voltage VH as a condition for executing the resolver origin learning and the rotation speed of the motor are limited to a narrow range, the probability that the origin learning can be performed is reduced, and the motor rotor There is a possibility that the frequency of performing the origin correction decreases, or the origin correction of the rotor of the motor cannot be performed. Therefore, in the motor control device according to the present embodiment, a low-accuracy region indicated by thin shading in FIG. The low accuracy region is a range defined by a combination of the system voltage VH and the motor rotation speed over a wider range than the high accuracy region. In the motor control device according to the present embodiment, since the resolver origin learning is executed even when the system voltage VH and the motor rotation speed correspond to such a low accuracy region, the origin learning frequency is reduced. Can be suppressed.

  In addition, in the motor control device according to the present embodiment, when the origin learning is executed at the system voltage VH corresponding to the low accuracy region and the rotation speed of the motor, the origin learning is performed again with high accuracy. Retry learning. Thereby, the fall of the learning precision accompanying the increase in learning frequency can be suppressed. Note that, when the system voltage VH and the rotation speed of the motor do not correspond even to the low accuracy region, the origin learning of the resolver is not executed. Thereby, in the motor control device according to the present embodiment, even when trying to improve the learning accuracy by executing the origin learning of the resolver in a state where the system voltage VH and the rotational speed of the motor are in a predetermined range, The frequency of origin learning can be secured.

  On the other hand, as shown in FIG. 4B, in the case where only the high accuracy region is provided without providing the low accuracy region as described above, the system voltage VH and the motor rotation speed are relatively narrow. Since the resolver origin learning is performed only when applicable, the frequency of origin learning is reduced, the frequency of origin correction of the motor rotor is reduced, or the origin correction of the motor rotor is performed. There is a risk that it may become impossible.

  It should be noted that the high accuracy region and the low accuracy region shown in the coordinate system consisting of the voltage (system voltage VH) axis and the motor rotation speed axis shown in FIG. The high accuracy region and the low accuracy region can be appropriately determined according to the configuration, specifications, characteristics, and the like of each motor control device and motor.

  Incidentally, the accuracy of the resolver origin learning in the motor control device is influenced not only by the motor rotation speed and the system voltage VH, which is the input voltage to the inverter, but also by various other factors. Therefore, in order to perform the origin learning of the resolver with higher accuracy, it is desirable to appropriately control various factors that affect the learning accuracy and perform the origin learning in a situation where higher learning accuracy can be obtained. . As a further example of such factors, for example, the carrier frequency of the inverter can be cited. In this case, the origin of the resolver can be learned with higher accuracy by controlling the carrier frequency of the inverter to be an appropriate frequency.

  Specifically, when the carrier frequency of the inverter is too low, it becomes difficult to secure the control frequency of the command voltage necessary to achieve the motor rotation speed when performing the origin learning, and the motor rotation is reduced. There is a possibility that the motor speed becomes stable and the rotation speed of the motor cannot be kept constant. Thus, when the rotation of the motor becomes unstable, the accuracy of the resolver origin learning also decreases. Therefore, when performing the resolver origin learning, a value exceeding the minimum carrier frequency necessary to maintain the motor rotation speed is required. It is desirable to set the frequency as the target frequency.

  On the other hand, as disclosed in another application by the present inventor, as the carrier frequency of the inverter increases, the number of dead times included in one period of the electrical angle of the rotor increases, so that dead time is provided. The present inventors have found that the decrease in the accuracy of the resolver origin learning due to the above becomes more significant as the carrier frequency of the inverter becomes higher.

  Here, the influence of the carrier frequency of the inverter on the decrease in the accuracy of the resolver origin learning caused by providing the dead time will be described in detail with reference to the accompanying drawings. FIG. 5 is a schematic diagram illustrating the influence of the carrier frequency of the inverter on the decrease in the accuracy of the resolver origin learning caused by providing the dead time as described above. As described above with reference to FIG. 2, in the duty ratio of the upper arm and the lower arm of the inverter, by providing a dead time, the command duty ratio calculated based on the target output voltage, and the actual upper arm and There is a difference between the duty ratio (actual duty ratio) when the lower arm is turned on / off.

  In the above, as shown in FIG. 5 (1), when the carrier frequency of the inverter is low, the number of dead times (shaded portions in FIG. 5) included in one cycle of the electrical angle of the rotor is small, and the dead time is reduced. By providing the difference, the difference between the command duty ratio and the actual duty ratio is small. On the other hand, as shown in FIG. 5 (2), when the carrier frequency of the inverter is high, the dead time (shaded portion in FIG. 5) included in one cycle of the electrical angle of the rotor is large and the dead time is provided. As a result, there is a large difference between the command duty ratio and the actual duty ratio. As described above, as the carrier frequency of the inverter increases, the accuracy of the resolver origin learning caused by providing the dead time becomes more significant.

  Therefore, in order to execute the origin learning of the resolver with higher accuracy, the carrier frequency of the inverter becomes a frequency equal to or higher than the minimum frequency required to achieve the rotation speed of the motor when performing the origin learning. The frequency at which the accuracy of the resolver origin learning decreases to an unacceptable level considering the decrease in the accuracy of the resolver origin learning due to the dead time as described above. It is desirable to control so as not to reach. More preferably, for example, it is desirable to control the carrier frequency of the inverter so as to be the minimum necessary frequency corresponding to the rotational speed of the motor when performing the origin learning of the resolver.

  In this case, when the carrier frequency and the number of rotations of the motor fall within a range (first range) (high accuracy region) in which a desired learning accuracy is obtained, which is defined by a combination of the carrier frequency and the number of rotations of the motor. By executing the resolver origin learning only, the resolver origin learning can be performed with higher accuracy. However, as described above, if the number of parameters used as the condition for executing the resolver origin learning is increased, the probability that all the parameters fall within a suitable range for executing the resolver origin learning decreases. As a result, the frequency at which the resolver origin learning is performed decreases, and the frequency at which the motor rotor origin correction is performed may decrease, or the motor rotor origin correction may not be performed.

  Therefore, even in such a case, as described above, the first range (high accuracy region) and the first range suitable for achieving desired learning accuracy for these parameters (carrier frequency and motor rotation speed). If the second range (low accuracy region) is defined and the parameters do not correspond to the high accuracy region but fall within the low accuracy region, the resolver origin learning is performed and the origin learning is performed again. It is desirable to suppress the decrease in the accuracy of the resolver origin learning while ensuring the frequency of the resolver origin learning.

That is, the third embodiment of the present invention
A motor control device according to the first embodiment of the present invention,
If the resolver origin learning is allowed,
Change the target frequency of the carrier frequency of the inverter to manipulate the carrier frequency,
When the carrier frequency and the rotation speed of the motor are within a first range defined by a combination of the carrier frequency and the rotation speed of the motor, the origin learning of the resolver is performed,
A second range defined by a combination of the carrier frequency and the motor rotational speed, the carrier frequency and the motor rotational speed not entering the first range and extending over a wider range than the first range. When entering, while performing the origin learning of the resolver, repeat the process after the learning permission determination,
If the carrier frequency and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.
It is a motor control device.

  As described above, in the motor control device according to this embodiment, when the origin learning of the resolver is permitted, the target frequency of the carrier frequency of the inverter is changed to operate the carrier frequency. Thereby, the origin of the resolver is learned by controlling the carrier frequency to be the target frequency. As a result, in the motor control device according to the present embodiment, the resolver origin learning can be performed with higher accuracy by controlling the carrier frequency to be an appropriate frequency.

  However, as described above, if the range of the carrier frequency and the number of rotations of the motor is limited as a condition for executing the origin learning of the resolver as described above, the probability that the origin learning can be performed is reduced, and the rotor of the motor There is a possibility that the frequency of performing the origin correction will decrease, or the origin correction of the rotor of the motor cannot be performed.

Therefore, in the motor control device according to this embodiment, as described above,
If the resolver origin learning is allowed,
Change the target frequency of the carrier frequency of the inverter to manipulate the carrier frequency,
When the carrier frequency and the rotation speed of the motor are within a first range defined by a combination of the carrier frequency and the rotation speed of the motor, the origin learning of the resolver is performed,
A second range defined by a combination of the carrier frequency and the motor rotational speed, the carrier frequency and the motor rotational speed not entering the first range and extending over a wider range than the first range. When entering, while performing the origin learning of the resolver, repeat the process after the learning permission determination,
When the carrier frequency and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.

  As described above, in the motor control device according to the present embodiment, the determination as to whether or not the resolver origin learning is performed is made based on the carrier frequency of the inverter and the rotational speed of the motor. More specifically, in the motor control apparatus according to the present embodiment, a range (first range) of parameters (carrier frequency and motor rotation speed) that can execute the origin learning of the resolver with desired learning accuracy. ) Performs origin learning when these parameters are applicable. In addition, even when these parameters do not fall within the above range (first range), the learning accuracy is low compared to the above range (first range), but these are within a wide range (second range). The origin learning is also executed when the parameters (carrier frequency and motor rotation speed) are applicable. Thereby, the frequency of origin learning can be increased.

  However, the accuracy of the origin learning is lowered only by controlling as described above. Therefore, in the motor control device according to the present embodiment, when the resolver origin learning is executed in a case where these parameters do not fall within the first range but fall within the second range, high accuracy is again obtained. Retry origin learning to perform origin learning at. Thereby, the fall of the learning precision accompanying the increase in learning frequency can be suppressed. If these parameters do not even fall within the second range, origin learning is not executed. Thereby, in the motor control device according to the present embodiment, even when trying to improve the learning accuracy by executing the resolver origin learning in a state where the carrier frequency and the rotation speed of the motor are in a predetermined range, The frequency of learning can be secured.

  In the first range and the second range defined by the combination of the carrier frequency and the motor rotation speed, the upper limit value and / or the lower limit value of any one parameter is a parameter other than the one parameter. It may be determined independently of the value. Alternatively, in the first range and the second range defined by the combination of the carrier frequency and the rotation speed of the motor, the upper limit value and / or the lower limit value of any one parameter is a parameter other than the one parameter. It may change depending on the value.

  By the way, in the above, the first range (high accuracy region) and the second defined respectively by the motor rotation number alone, the combination of the system voltage VH and the motor rotation strike, and the combination of the carrier frequency and the motor rotation strike. Embodiments for determining the range (low accuracy region) have been described. However, the first range (high accuracy region) and the second range (low accuracy region) may be defined by a combination of the system voltage VH, the carrier frequency, and the rotational speed of the motor.

That is, the fourth embodiment of the present invention is
A motor control device according to the first embodiment of the present invention,
If the resolver origin learning is allowed,
Manipulating the system voltage VH by changing the target voltage value of the system voltage VH that is an input voltage to the inverter, and changing the target frequency of the carrier frequency of the inverter to manipulate the carrier frequency,
When the carrier frequency, the system voltage VH, and the rotation speed of the motor fall within a first range defined by a combination of the carrier frequency, the system voltage VH, and the rotation speed of the motor, Perform origin learning,
The carrier frequency, the system voltage VH, and the rotation speed of the motor, in which the carrier frequency, the system voltage VH, and the rotation speed of the motor do not fall within the first range but are wider than the first range. When entering the second range defined by the combination with and performing the origin learning of the resolver, repeat the process after the learning permission determination,
When the carrier frequency, the system voltage VH, and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.
It is a motor control device.

  As described above, in the motor control device according to the present embodiment, when the origin learning of the resolver is permitted, the target voltage value of the system voltage VH that is the input voltage to the inverter is changed to change the system voltage. The VH is operated, and the carrier frequency is changed by changing the target frequency of the carrier frequency of the inverter. Thus, the system voltage VH is controlled to be the target voltage value, and the carrier frequency is controlled to be the target frequency to perform the origin learning of the resolver. As a result, in the motor control device according to the present embodiment, the system voltage VH is controlled to be an appropriate voltage value, and the carrier frequency is controlled to be an appropriate frequency, thereby further improving the accuracy. Resolver origin learning can be performed.

  However, as described above, if the range of the system voltage VH, the carrier frequency, and the motor rotation speed is limited as a condition for executing the resolver origin learning as described above, the probability that the origin learning can be performed is reduced. There is a possibility that the frequency of correcting the origin of the rotor of the motor may be reduced or the origin of the rotor of the motor cannot be corrected.

Therefore, in the motor control device according to this embodiment, as described above,
If the resolver origin learning is allowed,
When the carrier frequency, the system voltage VH, and the rotation speed of the motor fall within a first range defined by a combination of the carrier frequency, the system voltage VH, and the rotation speed of the motor, Perform origin learning,
The carrier frequency, the system voltage VH, and the rotation speed of the motor, in which the carrier frequency, the system voltage VH, and the rotation speed of the motor do not fall within the first range but are wider than the first range. When entering the second range defined by the combination with and performing the origin learning of the resolver, repeat the process after the learning permission determination,
If the carrier frequency, the system voltage VH, and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.

  As described above, in the motor control device according to the present embodiment, the determination as to whether or not the resolver origin learning is performed is based on the system voltage VH that is the input voltage to the inverter, the carrier frequency of the inverter, and the rotation of the motor. Done based on the number. More specifically, in the motor control device according to the present embodiment, ranges of parameters (system voltage VH, carrier frequency, and motor rotation speed) that can execute resolver origin learning with desired learning accuracy. When these parameters correspond to (first range), origin learning is executed. In addition, even when these parameters do not fall within the above range (first range), the learning accuracy is low compared to the above range (first range), but these are within a wide range (second range). The origin learning is also executed when the parameters (system voltage VH, carrier frequency, and motor rotation speed) are applicable. Thereby, the frequency of origin learning can be increased.

  However, the accuracy of the origin learning is lowered only by controlling as described above. Therefore, in the motor control device according to the present embodiment, when the resolver origin learning is executed in a case where these parameters do not fall within the first range but fall within the second range, high accuracy is again obtained. Retry origin learning to perform origin learning at. Thereby, the fall of the learning precision accompanying the increase in learning frequency can be suppressed. If these parameters do not even fall within the second range, origin learning is not executed. Thereby, in the motor control apparatus according to the present embodiment, the learning accuracy is improved by performing the origin learning of the resolver in a state where the system voltage VH, the carrier frequency, and the rotation speed of the motor are in a predetermined range. Even in this case, the frequency of origin learning can be ensured.

  In the first range and the second range defined by the combination of the system voltage VH, the carrier frequency, and the motor speed, the upper limit value and / or the lower limit value of any one parameter is the one parameter. It may be determined independently from the values of other parameters. Alternatively, in the first range and the second range defined by the combination of the system voltage VH, the carrier frequency, and the rotation speed of the motor, the upper limit value and / or the lower limit value of any one parameter is the one parameter It may change depending on the value of a parameter other than.

  Here, the origin learning of the resolver executed in the motor control device according to the present embodiment will be described in more detail with reference to the attached drawings. FIG. 6 is a flowchart showing the flow of various processes in an example of the resolver origin learning executed in the motor control device according to one embodiment of the present invention as described above. As shown in FIG. 6, in the motor control device according to the present embodiment, when a routine for performing the origin learning of the resolver is started, learning permission determination is first performed in step S110. Specifically, in step S110, it is determined whether or not the rotational speed of the motor is within a predetermined range and the required torque for the motor is 0 (zero) (torque command is 0 (zero)). Determined.

  When it is determined in step S110 that the condition for permitting the origin learning of the resolver is not satisfied (learning is not permitted) (step S110 = No), the routine ends. On the other hand, if it is determined in step S110 that the condition for permitting the origin learning of the resolver is satisfied (learning is permitted) (step S110 = Yes), the input voltage to the inverter is determined in the next step S120. The target voltage value of a certain system voltage VH is changed, and the system voltage VH is operated so as to reach the target voltage value in step S130. Furthermore, in step S140, the target frequency of the carrier frequency of the inverter is changed, and in step S150, the inverter is operated so that the carrier frequency reaches the target frequency.

  Thus, when the operation of the system voltage VH and the carrier frequency is completed, in the next step S160, each parameter (system voltage VH, carrier frequency, and motor rotation speed) corresponds to the second range (low accuracy region). It is determined whether or not. If it is determined in step S160 that each parameter does not fall within the second range (low-precision region) (step S160 = No), the routine ends. On the other hand, when it is determined in step S160 that each parameter corresponds to the second range (low accuracy region) (step S160 = Yes), each parameter is in the second range (high accuracy region) in the next step S170. ) Is determined.

  If it is determined in step S170 that each parameter falls within the first range (high accuracy region) (step S170 = Yes), the resolver origin learning is executed with high accuracy in the next step S180. On the other hand, when it is determined in step S170 that each parameter does not fall within the first range (high accuracy region) (step S160 = No), the origin learning of the resolver is performed with high accuracy in the next step S190. However, the routine returns to step S110 again, and origin learning is executed again.

  As described above, according to the motor control device of the present embodiment, learning is performed by performing resolver origin learning in a state where the system voltage VH, the carrier frequency, and the rotation speed of the motor are within predetermined ranges as described above. Even when trying to increase accuracy, the frequency of origin learning can be secured.

  By the way, in the motor control device according to the present invention including various embodiments including the above-described several embodiments, the origin of the resolver is learned (with high accuracy) in a state where each parameter falls within the first range. Can be used as the resolver offset value, and can be used, for example, for the origin correction of the resolver. On the other hand, when the resolver origin learning cannot be executed with high accuracy in a state where each parameter falls within the first range, the learning result is adopted as the resolver offset value. It can be used for origin correction.

That is, the fifth embodiment of the present invention
A motor control device according to any one of the first to fourth embodiments of the present invention,
If the origin learning can be executed in a state corresponding to the first range, the offset value of the resolver is specified based on the result of the origin learning executed in the state corresponding to the first range. And
If the origin learning cannot be executed in the state corresponding to the first range, the resolver offset value is calculated based on the result of the origin learning performed in the state corresponding to the second range. Identify,
It is a motor control device.

  As described above, in the motor control device according to the present embodiment, when origin learning can be executed in a state corresponding to the first range, the motor control device is executed in a state corresponding to the first range. If the resolver offset value is specified based on the result of the origin learning performed and the origin learning cannot be executed in a state corresponding to the first range, it corresponds to the second range. The resolver offset value is specified based on the result of the origin learning performed in the state.

  Thereby, in the motor control apparatus according to the present embodiment, not only when the resolver origin learning can be executed (with high accuracy) in a state where each parameter falls within the first range, Even when resolver origin learning cannot be executed (with high accuracy) in a state corresponding to the first range, the frequency of resolver origin learning is ensured, and the result of origin learning is The offset value can be used, for example, for the origin correction of a resolver.

  By the way, although the resolver origin learning could not be executed in a state where each parameter corresponds to the first range (with high accuracy), in a state where each parameter corresponds to the second range (to low accuracy) When the resolver origin learning can be executed a plurality of times, an average value of these learning results can be calculated as an offset value of the resolver and used for, for example, correction of the resolver origin.

That is, the sixth embodiment of the present invention
A motor control apparatus according to the fifth embodiment of the present invention,
When origin learning cannot be executed in a state corresponding to the first range, an offset value of the resolver is used as an average value of results of origin learning performed in a state corresponding to the second range. calculate,
It is a motor control device.

  If the origin learning cannot be executed in the state corresponding to the first range, the resolver offset value is obtained as an average value of the results of the origin learning executed in the state corresponding to the second range. Is calculated, a weighted average may be obtained by assigning a weight to each of a plurality of learning results. At this time, the weights assigned to the individual learning results can be determined based on, for example, the degree of deviation from a parameter value or a combination of a plurality of parameter values that can achieve the highest accuracy in the first range. it can.

  As described above, in the motor control device according to the present invention including the various embodiments described above, parameters such as the system voltage VH, the carrier frequency, and the rotational speed of the motor are provided. Even when trying to improve the learning accuracy by executing the origin learning of the resolver in a state where it falls within the predetermined range, the origin learning of the resolver can be performed with high accuracy while ensuring the frequency of the origin learning.

  In the foregoing, for the purpose of illustrating the present invention, several embodiments having specific configurations and combinations of execution procedures have been described, but the scope of the present invention is limited to these exemplary embodiments. Instead, modifications can be appropriately made within the scope of the matters described in the claims and the specification.

Claims (6)

Based on the rotation angle of the rotor of the motor detected by the resolver, the rotation of the motor is controlled by controlling the voltage value and phase of the command voltage from the inverter including the switching element to the motor,
When the motor is rotating and the required torque for the motor is 0 (zero), the origin of the resolver is learned,
Providing a dead de time is a period electrical conduction is interrupted at the same time of upper and lower arms of the inverter,
In the motor control device,
If the required torque speed relative and the motor Ri range near a predetermined of the motor is 0 (zero) to allow starting learned of the resolver, the range the rotational speed of the motor is predetermined If the required torque for the motor is not within 0 (zero), the learning permission determination is performed to prohibit the origin learning of the resolver ,
When the starting learned of the resolver is permitted by the learning permission judgment,
Rotational speed of the motor, when entering the first range defined by the rotational speed of the motor, perform home learning of the resolver,
Rotational speed of the motor, said when entering the second range defined by the rotational speed of the motor up to and wider than the first range not enter the first range, the starting learned of the resolver And repeating the process after the learning permission determination,
When the rotation speed of the motor does not fall within the second range, without starting learned of the resolver,
If the learning origin determination of the resolver is prohibited by the learning permission determination, the origin learning of the resolver is not performed .
Motor control device. The motor control device according to claim 1,
If the resolver origin learning is allowed,
The system voltage VH is manipulated by changing the target voltage value of the system voltage VH, which is the input voltage to the inverter,
When the system voltage VH and the rotation speed of the motor fall within a first range defined by a combination of the system voltage VH and the rotation speed of the motor, the origin learning of the resolver is performed,
The system voltage VH and the rotation speed of the motor are defined by a combination of the system voltage VH and the rotation speed of the motor that do not fall within the first range and extend over a wider range than the first range. When entering the range of, the origin of the resolver is learned, and the process after the learning permission determination is repeated,
When the command voltage and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.
Motor control device. The motor control device according to claim 1,
If the resolver origin learning is allowed,
Change the target frequency of the carrier frequency of the inverter to manipulate the carrier frequency,
When the carrier frequency and the rotation speed of the motor are within a first range defined by a combination of the carrier frequency and the rotation speed of the motor, the origin learning of the resolver is performed,
A second range defined by a combination of the carrier frequency and the motor rotational speed, the carrier frequency and the motor rotational speed not entering the first range and extending over a wider range than the first range. When entering, while performing the origin learning of the resolver, repeat the process after the learning permission determination,
If the carrier frequency and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.
Motor control device. The motor control device according to claim 1,
If the resolver origin learning is allowed,
Manipulating the system voltage VH by changing the target voltage value of the system voltage VH that is an input voltage to the inverter, and changing the target frequency of the carrier frequency of the inverter to manipulate the carrier frequency,
When the carrier frequency, the system voltage VH, and the rotation speed of the motor fall within a first range defined by a combination of the carrier frequency, the system voltage VH, and the rotation speed of the motor, Perform origin learning,
The carrier frequency, the system voltage VH, and the rotation speed of the motor, in which the carrier frequency, the system voltage VH, and the rotation speed of the motor do not fall within the first range but are wider than the first range. When entering the second range defined by the combination with and performing the origin learning of the resolver, repeat the process after the learning permission determination,
When the carrier frequency, the system voltage VH, and the rotation speed of the motor do not fall within the second range, the origin learning of the resolver is not performed.
Motor control device. The motor control device according to any one of claims 1 to 4,
If the origin learning can be executed in a state corresponding to the first range, the offset value of the resolver is specified based on the result of the origin learning executed in the state corresponding to the first range. And
If the origin learning cannot be executed in the state corresponding to the first range, the resolver offset value is calculated based on the result of the origin learning performed in the state corresponding to the second range. Identify,
Motor control device. The motor control device according to claim 5,
When origin learning cannot be executed in a state corresponding to the first range, an offset value of the resolver is used as an average value of results of origin learning performed in a state corresponding to the second range. calculate,
Motor control device.

Images