JP2015169464A - Motor rotation angle detection device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a motor rotation angle detection device which is capable of suppressing the reduction in resolution and accuracy in the case that two signals cannot be simultaneously subjected to A/D conversion.SOLUTION: A rotation angle detection device 1 includes a resolver 3 for outputting first and second signals corresponding to a rotation angle of an electric motor M and detects a rotation angle of the electric motor. The rotation angle detection device 1 detects the rotation angle of the electric motor on the basis of a voltage at a peak position of amplitude of the first signal outputted from the resolver and a voltage at a timing shifted from a peak position of amplitude of a second signal by a prescribed time. In the case that the first and second signals cannot be simultaneously subjected to A/D conversion, both of the first and second signals outputted from the resolver can be handled as acquisition values of peak positions by performing correction processing on the voltage at the timing shifted by the prescribed time, and the reduction in resolution and accuracy can be suppressed.

Description

  The present invention relates to a rotation angle detection device that detects a rotation angle of a motor using a resolver.

In an on-vehicle electric actuator, a permanent magnet brushless motor may be used in terms of efficiency in motor driving, and a method of detecting a rotation angle of the motor with a resolver is known. In this method, an excitation signal is input to the resolver, and the rotation angle of the motor is calculated based on a sine wave signal and a cosine wave signal output from the resolver.
Here, when A / D conversion of two signals cannot be performed at the same time (two A / D conversion functions cannot be used), for example, measurement is performed with a period shifted as in Patent Document 1, or Patent Document As shown in Fig. 2, measurement is performed by combining timings that theoretically have the same value within one period of the signal waveform.

JP 2011-223826 A JP 2000-39337 A

However, in the former technique, the timing for performing the A / D conversion of the two signals greatly deviates, and the calculation accuracy of the rotation angle of the motor decreases. On the other hand, in the latter technique, since the voltage is acquired before and after the maximum amplitude of the excitation signal waveform, the timing with the best resolution and accuracy is removed.
For this reason, the former causes a decrease in accuracy during transient tracking, and the latter has a problem that the detection timing with the best resolution and accuracy cannot be used.

  The present invention has been made in view of the circumstances as described above, and the object of the present invention is to provide a rotation angle of a motor that can suppress degradation in resolution and accuracy when A / D conversion of two signals cannot be performed simultaneously. It is to provide a detection device.

  In the motor rotation angle detection device of the present invention, one of the two signals output from the resolver is acquired at the peak position, and the other is acquired at a position shifted from the peak position for a predetermined time, and the rotation angle of the motor is based on these output values. It was made to detect.

  According to the present invention, when the A / D conversion of two signals output from the resolver cannot be performed at the same time, the correction is performed on the output value obtained by shifting the peak position by a predetermined time, thereby outputting from the resolver. Both of the two signals to be processed can be handled as the acquired values of the peak position, and the decrease in resolution and accuracy can be suppressed.

1 is a block diagram illustrating a schematic configuration of a motor rotation angle detection device according to an embodiment of the present invention. It is a timing chart which shows the 1st detection operation in the rotation angle detection apparatus shown in FIG. 3 is a flowchart illustrating a correction-side signal acquisition processing operation in FIG. 2. 3 is a flowchart illustrating a signal acquisition processing operation on a peak side in FIG. 2. It is a timing chart which shows the 2nd detection operation in the rotation angle detection apparatus shown in FIG. It is a flowchart which shows the signal acquisition processing operation of the correction | amendment side in FIG. 6 is a flowchart showing a signal acquisition processing operation on the peak side in FIG. 5. FIG. 6 is a timing chart showing a third detection operation in the rotation angle detection device shown in FIG. 1. FIG. It is a flowchart which shows the signal acquisition processing operation of the correction | amendment side in FIG. 9 is a flowchart illustrating a signal acquisition processing operation on the peak side in FIG. 8. It is a timing chart which shows the 4th detection operation in the rotation angle detection apparatus shown in FIG. It is a flowchart which shows the signal acquisition processing operation of the correction | amendment side in FIG. 12 is a flowchart showing a signal acquisition processing operation on the peak side in FIG. It is a schematic block diagram at the time of applying the rotation angle detection apparatus of the motor which concerns on embodiment of this invention to VCR. It is a wave form diagram for demonstrating the relationship between the rotation angle of the motor in VCR shown in FIG. 14, and the voltage value of the sine wave signal and cosine wave signal which are output from a resolver.

Embodiments of the present invention will be described below with reference to the drawings.
[First Embodiment]
FIG. 1 shows a schematic configuration of a motor rotation angle detection device according to an embodiment of the present invention. The rotation angle detection device 1 receives an excitation signal from the motor control device 2, outputs a sine wave signal and a cosine wave signal corresponding to the rotation angle of the motor (electric motor) M, and outputs from the resolver 3. An A / D converter 4 for A / D converting the sine wave signal and the cosine wave signal, and an arithmetic unit 5 for calculating the arc tangent by calculating the digital data output from the A / D converter 4 It is comprised including.

  The resolver 3 is attached to the rotor of the motor M, and rotates as a rotor 3a integrally with the motor M, an excitation coil 3b to which an excitation signal is supplied from the motor control device 2, and an excitation coil via the rotor 3a. And a pair of detection coils 3c and 3d that generate an electromotive force by the magnetic force of 3b. Here, the pair of detection coils 3c and 3d are arranged with a phase difference of 90 ° in order to generate AC voltages (resolver signals) of sine waves and cosine waves.

  When the rotor 3a of the resolver 3 rotates with the rotation of the motor M, a sine wave signal and a cosine wave signal are output from the detection coils 3c and 3d, respectively, according to the rotation angle of the rotor 3a. The rotation angle of the motor M is determined by inputting the voltage values of the sine wave signal and the cosine wave signal to the A / D converter 4 and performing A / D conversion to convert them into digital data. Calculated by taking Arctan). The calculated rotation angle is input to the motor control device 2 to feedback control the motor M.

  The first detection operation of the rotation angle detection apparatus 1 utilizes the fact that the period and amplitude of the excitation signal input to the resolver 3 are constant as shown in the timing chart of FIG. That is, since the period and amplitude of the excitation signal are constant, the timing at which the amplitude ratio (α) is constant with respect to the maximum amplitude of the excitation signal is known. Therefore, the voltage value of the sine wave signal is detected at a timing before the peak position of the excitation signal (β), A / D conversion is performed, correction processing is performed from the ratio to the maximum amplitude, and the amplitude of the excitation signal is determined. A correction value approximated to the maximum value (peak position) is used after A / D conversion. On the other hand, the cosine wave signal is used after detecting the voltage value at the peak position of the excitation signal and performing A / D conversion. The rotation angle of the motor M is calculated by taking the arc tangent of the voltage value acquired from these sine wave signal and cosine wave signal.

  As an example, when it is desired to detect a voltage with an amplitude ratio α = 0.9 (90%) with respect to the amplitude at times t1, t5, t9, t13, and t17, which are peak positions of the excitation signal, the period T of the excitation signal is Since it is constant (for example, T = 100 μs), it can be seen that 92.77 μs has elapsed from the previous peak position of the excitation signal. Therefore, the voltage value of the sine wave signal when 92.77 μs has elapsed from the previous peak position is multiplied by the reciprocal of 0.9 (= 1 ÷ 0.9 = 1.11) and used as a correction value. .

  Therefore, the voltage acquisition processing timing on the correction side (side of the sine wave signal) is a predetermined time β (7.23 μs in this example) with respect to times t1, t5, t9, t13, and t17, which are peak positions of the excitation signal. The previous times t0, t4, t8, t12, and t16. As described above, the voltage acquisition processing timing is equivalent to the acquisition of the voltage value of the sine wave signal at the times t1, t5, t9, t13, and t17 of the peak position of the excitation signal while the predetermined time β is before the peak position. become. On the other hand, the voltage acquisition processing timing on the peak side (cosine wave signal side) is times t1, t5, t9, t13, and t17 of the peak position of the excitation signal. The motor angle update timing is the voltage acquisition processing timing on the peak side, and is time t1, t5, t9, t13, t17.

Next, the processing operation of the acquired voltage value will be described in detail with reference to the flowcharts of FIGS. FIG. 3 is a flowchart showing the voltage acquisition processing operation on the correction side in FIG. In step S1, for example, the voltage value of the sine wave signal is acquired at the timing of time t0, and A / D conversion is performed. Here, the voltage value range of the sine wave signal is −1 to +1.
In step S2, the voltage value of the acquired sine wave signal is set to Vsin1, and in the next step S3, the voltage value Vsin1 is corrected so as to correspond to the voltage value of the sine wave signal at the timing of time t1. In this correction processing, calculation of “Vsin1hosei = Vsin1 × correction coefficient” is performed. The correction coefficient is 1.11 when the amplitude ratio α = 0.9.

Similarly, a voltage value corresponding to the voltage value of the sine wave signal at timings t5, t9, t13, and t17 is obtained by acquiring the voltage value of the sine wave signal at timings t4, t8, t12, and t16. Calculate the value.
Since the amplitude and period of the excitation signal are always constant in this way, the voltage value of the sine wave signal is acquired and A / D converted at a timing shifted by a predetermined time with respect to the time when the excitation signal has the maximum amplitude. And correcting the digital data with the amplitude ratio with respect to the maximum amplitude, the voltage value of the sine wave signal approximate to that when the excitation signal has the maximum amplitude is obtained.

FIG. 4 is a flowchart showing the voltage acquisition processing operation on the peak side. In step S4, for example, the voltage value of the cosine wave signal is acquired at the timing of time t1, and A / D conversion is performed. The voltage value range of this cosine wave signal is also −1 to +1. In step S5, the voltage value of the acquired cosine wave signal is set to Vcos1, and the motor angle is calculated in the next step S6. The motor angle calculation is calculated from Vsin1hosei and Vcos1 using an arc tangent.
Thereafter, similarly, the voltage corresponding to the voltage value of the sine wave signal at the times t5, t9, t13, and t17 calculated by the correction process is acquired at the timing of the times t5, t9, t13, and t17. The motor angle is calculated using the value Vsin1hosei.

According to the above configuration, when A / D conversion of the voltage value acquired from the sine wave signal and cosine wave signal output from the resolver cannot be performed at the same time, correction processing is performed on the shifted side. Since both the two signal waveforms output from the resolver can be handled as the output value of the peak position of the excitation signal, it is possible to suppress a decrease in resolution and accuracy.
The first embodiment is effective when it is desired to reduce the processing load of the arithmetic device 5 because the update timing of the rotation angle of the motor is relatively small.
Also, since the voltage value of the sine wave signal is detected at a timing before a predetermined time with respect to the peak position of the excitation signal, the correction processing is calculated using the time β until the detection timing of the cosine wave signal. Can do.

  In the first embodiment, the voltage value of the sine wave signal is acquired at a timing shifted by a predetermined time with respect to the time point when the amplitude of the excitation signal becomes the maximum amplitude, and A / D conversion is performed to perform correction processing. However, as a matter of course, the voltage value of the cosine wave signal may be acquired, A / D converted, and corrected. In this case, the voltage value of the sine wave signal is acquired at the timing when the excitation signal has the maximum amplitude, and A / D conversion is performed.

[Second Embodiment]
FIG. 5 is a timing chart showing the second detection operation. In the second detection operation, the voltage values of the sine wave signal and the cosine wave signal are acquired at the upper and lower peak positions of the amplitude of the excitation signal to increase the angle update timing of the motor M, and the reference voltage is obtained to obtain the reference voltage of the motor. The rotation angle is detected.

  The voltage acquisition processing timing on the correction side (sine wave signal side) is a predetermined time β with respect to times t1, t3, t5, t7, t9, t11, t13, t15, and t17, which are the upper and lower peak positions of the excitation signal. The previous times t0, t2, t4, t6, t8, t10, t12, t14, and t16. As described above, while the voltage acquisition processing timing is a predetermined time β before the peak position, the correction processing is performed, so that the time t1, t3, t5, t7, t9, t11, t13, This is equivalent to acquiring the voltage value of the sine wave signal at t15 and t17. On the other hand, the voltage acquisition processing timing on the peak side (cosine wave signal side) is times t1, t3, t5, t7, t9, t11, t13, t15, t17, which are the upper and lower peak positions of the excitation signal. The motor angle update timing is a peak-side voltage acquisition processing timing, and is time t1, t3, t5, t7, t9, t11, t13, t15, t17.

For example, at the timing of time t4, the reference voltage is obtained from the average value of the sine wave signal between time t2 and t4. For example, at the timing of time t5, the reference voltage is calculated from the average value of the cosine wave signal between time t3 and t5. Ask. At the timing of time t8, the reference voltage is obtained from the average value of the sine wave signal between times t6 and t8, and at the timing of time t9, the reference voltage is obtained from the average value of the cosine wave signal between times t7 and t9. Similarly, at times t12 and t16, a reference voltage is obtained from the average values of the sine wave signals between times t10 and t12 and between t14 and t16, and at times t13 and t17, the time t11− The reference voltage is obtained from the average values of the cosine wave signals between t13 and t15-t17.
The update timing of the reference voltage is time t0, t1, t4, t5, t8, t9, t12, t13, t16, t17. Here, the reference voltage update flag FLAG_cen is “1”, that is, updated before time t1, between time t3 and t5, between time t7 and t9, between time t11 and t13, and between time t15 and t17.

Next, the processing operation of the acquired voltage will be described in detail with reference to the flowcharts of FIGS. FIG. 6 is a flowchart showing the voltage acquisition processing operation on the correction side in FIG. In step S11, the voltage value of the sine wave signal is acquired and A / D conversion is performed. The voltage value range of the sine wave signal is −1 to +1.
In step S12, it is determined whether it is a reference voltage update timing. This determination is made based on whether or not the reference voltage update flag FLAG_cen = 1. For example, when the flag FLAG_cen is “1” at the time t0, the voltage value of the acquired sine wave signal is set to Vsin2 (step S13), and correction processing is performed on the Vsin2 in the next step S14. In the correction process, the calculation of “Vsin2hosei = Vsin2 × correction coefficient” is performed. The correction coefficient is 1.11 when the amplitude ratio α = 0.9.

In step S15, the reference voltage Vscen is calculated. The reference voltage Vscen is an average value with the previous correction value Vsin1hosei, that is, “Vscen = (Vsin1hosei + Vsin2hosei) / 2”. In step S16, a difference voltage Vs3ho (Vs3ho = Vsin2hosei−Vscen) between the correction value Vsin2hosei and the reference voltage Vscen is calculated.
The operation for acquiring the voltage value of the sine wave signal at times t4, t8, t12, and t16 is the same.

On the other hand, if the flag FLAG_cen is “0” at the timing t2, which is not the update timing of the reference voltage in step S12, the voltage value of the acquired sine wave signal is set to Vsin1 (step S17). On the other hand, correction processing is performed in the next step S18. In the correction process, the calculation of “Vsin1hosei = Vsin1 × correction coefficient” is performed. The correction coefficient is 1.11 when the amplitude ratio α = 0.9. In step S19, a difference voltage Vs3ho (Vs3ho = Vsin1hosei−Vscen) between the correction value Vsin1hosei and the reference voltage Vscen is calculated.
The operation for acquiring the voltage value of the sine wave signal at times t6, t10, t14, and t18 is the same.

FIG. 7 is a flowchart showing the voltage acquisition processing operation on the peak side. In step S21, a cosine wave signal is acquired and A / D conversion is performed. The range of the cosine wave signal value is −1 to +1.
In step S22, it is determined whether it is a reference voltage update timing. This determination is made based on whether or not the reference voltage update flag FLAG_cen = 1. For example, when the flag FLAG_cen is “1” at time t1, the voltage value of the acquired cosine wave signal is set to Vcos2 (step S23), and the reference voltage Vccen is calculated for the Vcos2 in the next step S24. . The reference voltage Vccen is an average value with the previous acquired value Vcos1, that is, “Vccen = (Vcos1 + Vcos2) / 2”. In step S25, a difference voltage Vcos3 (Vcos3 = Vcos2-Vccen) between the voltage value Vcos2 and the reference voltage Vccen is calculated. Subsequently, the flag FLAG_cen is cleared to “0” (FLAG_cen = 0).
The subsequent operation for acquiring the voltage value of the cosine wave signal at times t5, t9, t13, and t17 is the same.

On the other hand, when it is determined in step S22 that it is not the update timing of the reference voltage, for example, when the flag FLAG_cen is “0” at the timing of time t3, the voltage value of the acquired cosine wave signal is set to Vcos1 (step S27). In the next step S28, "Vcos3 = Vcos1-Vccen" is calculated for this Vcos1. Subsequently, the flag FLAG_cen is set to “1” (FLAG_cen = 1).
The subsequent operation of acquiring the voltage value of the cosine wave signal at times t7, t11, t15, and t19 is the same.
In step S30, the motor angle is calculated. In this calculation, the motor angle is calculated using arc tangent from the difference voltage Vs3ho of the sine wave signal and the difference voltage Vcos3 of the cosine wave signal.

  According to the second detection operation as described above, both the two signal waveforms output from the resolver 3 can be handled as the output value of the peak position, similarly to the first detection operation, so that the resolution and accuracy are reduced. Can be suppressed. In addition, since the motor rotation angle update timing is greater than in the first detection operation, higher accuracy can be achieved.

  In the second embodiment, as in the first embodiment, the voltage value of the sine wave signal is acquired at a timing shifted by a predetermined time with respect to the time point when the excitation signal has the maximum amplitude, and A / D conversion is performed. However, the correction processing may be performed by acquiring the voltage value of the cosine wave signal and performing A / D conversion. In this case, the voltage value of the sine wave signal is acquired at the timing when the excitation signal has the maximum amplitude, and A / D conversion is performed.

[Third Embodiment]
FIG. 8 is a timing chart showing the third detection operation. In the third detection operation, a signal acquired at the timing of the peak position of the excitation signal is alternately acquired as a cosine wave signal and a sine wave signal.
The voltage acquisition processing timing of the sine wave signal is time t0, t5, t8, t13, t16, and the voltage acquisition processing timing of the cosine wave signal is time t1, t4, t9, t12, t17. Here, the voltage values acquired on the correction side are sine wave signals at times t0, t8, and t16, and cosine wave signals at times t4 and t12. The voltage value acquired as the peak side is a cosine wave signal at times t1, t9, and t17, and a sine wave signal at times t5 and t13. The motor angle update timing is the voltage acquisition processing timing on the peak side, and is time t1, t5, t9, t13, t17. The correction-side update flag FLAG_hosei is “1” between times t1 and t5, between times t9 and t13, and after time t17.

Next, the processing operation of the acquired voltage will be described in detail with reference to the flowcharts of FIGS. FIG. 9 is a flowchart showing the voltage acquisition processing operation on the correction side in FIG. In step S31, it is determined whether the correction target is a sine wave signal. For this determination, the correction side update flag FLAG_hosei is used. If FLAG_hosei = 0, a sine wave signal is acquired, and if FLAG_hosei = 1, a cosine wave signal is acquired. For example, at the timing of time t0, the voltage value of the sine wave signal is acquired and A / D conversion is performed (step S32). The voltage value range of the sine wave signal is −1 to +1.
In step S33, the voltage value of the acquired sine wave signal is set to Vsin1, and correction processing is performed on this Vsin1 in the next step S34. In the correction process, the calculation of “Vsin1hosei = Vsin1 × correction coefficient” is performed. The correction coefficient is 1.11 when the amplitude ratio α = 0.9. In step S35, the voltage Vsin1hosei obtained by the correction process is set as the peak value Vsinpeak of the sine wave signal.

On the other hand, if FLAG_hosei = 1 in step S31, a cosine wave signal is acquired, for example, at the timing of time t4, and A / D conversion is performed (step S36). The voltage value range of the cosine wave signal is −1 to +1.
In step S37, the voltage value of the acquired cosine wave signal is set to Vcos1, and correction processing is performed on this Vcos1 in the next step S38. In the correction process, the calculation of “Vcos1hosei = Vcos1 × correction coefficient” is performed. The correction coefficient is 1.11 when the amplitude ratio α = 0.9. In step S39, the voltage Vcos1hosei obtained by the correction process is set as the peak value Vcospeak of the cosine wave signal.
At subsequent times t8, t12, and t16, the voltage values of the sine wave signal, the cosine wave signal, and the sine wave signal are acquired and subjected to A / D conversion, and then correction processing is performed.

FIG. 10 is a flowchart showing the voltage acquisition processing operation on the peak side. In step S41, it is determined whether or not the correction target is a sine wave signal. For this determination, the correction side update flag FLAG_hosei is used. If FLAG_hosei = 0, the voltage value of the cosine wave signal is acquired, and if the flag FLAG_hosei = 1, the voltage value of the sine wave signal is acquired. For example, at the timing of time t1, the voltage value of the cosine wave signal is acquired and A / D conversion is performed (step S42). The range of the cosine wave signal value is −1 to +1.
In step S43, the voltage value of the acquired cosine wave signal is set to Vcos2, and this Vcos2 is set to the peak value Vcospeak of the cosine wave signal (step S44). In step S45, the flag FLAG_hosei is set to “1”.

On the other hand, if FLAG_hosei = 1 in step S41, a sine wave signal is acquired, for example, at the timing of time t5, and A / D conversion is performed (step S46). The voltage value range of the sine wave signal is −1 to +1.
In step S47, the voltage value of the acquired sine wave signal is set to Vsin2, and this Vsin2 is set to the peak value Vsinpeak of the sine wave signal. In step S49, the flag FLAG_hosei is cleared to “0”.
At times t9, t13, and t17, voltage values of the cosine wave signal, sine wave signal, and cosine wave signal are acquired and A / D conversion is performed.
In the next step S50, the motor angle is calculated. In this calculation, the motor angle is calculated using the arc tangent from the peak value Vsinpeak of the sine wave signal and the peak value Vcospeak of the cosine wave signal.

  According to the third detection operation as described above, both the two signal waveforms output from the resolver can be handled as the output value of the peak position, similarly to the first and second detection operations, so that the resolution is reduced. It is possible to suppress the deterioration of accuracy. In addition, by alternately switching the signal to be A / D converted at the timing of the peak position between the sine wave signal and the cosine wave signal, the accuracy deterioration can be distributed and evenly distributed as compared with the case where correction is performed with only one of the signals.

[Fourth Embodiment]
FIG. 11 is a timing chart showing the fourth detection operation. In the fourth detection operation, a correction process is performed on a cosine wave signal or a sine wave signal having a larger amplitude.
In the example shown in FIG. 11, the voltage acquisition processing timing of the cosine wave signal is time t0, t4, t8, t13, t17, and correction processing is performed on the voltage values acquired at time t0, t4, t8. The voltage acquisition processing timing of the sine wave signal is set to times t1, t5, t9, t12, and t16, and correction processing is performed on the voltage values acquired at times t12 and t16.
The motor angle update timing is the voltage acquisition processing timing on the peak side, and is time t1, t5, t9, t13, t17. The correction side update flag FLAG_hosei is “1” during the period up to time t9, and “0” after time t9.

  Next, the processing operation of the acquired voltage will be described in detail with reference to the flowcharts of FIGS. FIG. 12 is a flowchart showing the correction-side voltage acquisition processing operation in FIG. In step S51, it is determined whether the correction target is a sine wave signal. For this determination, the correction-side update flag FLAG_hosei is used. If FLAG_hosei = 0, the voltage value of the sine wave signal is acquired, and if FLAG_hosei = 1, the voltage value of the cosine wave signal is acquired.

For example, at the times t0, t4, and t8, the voltage value of the cosine wave signal is acquired and A / D conversion is performed (step S56). The voltage value range of the cosine wave signal is −1 to +1.
In step S57, the voltage value of the acquired cosine wave signal is set to Vcos1, and correction processing is performed on this Vcos1 in the next step S58. In the correction process, the calculation of “Vcos1hosei = Vcos1 × correction coefficient” is performed. The correction coefficient is 1.11 when the amplitude ratio α = 0.9. In step S59, Vcos1hosei obtained by the correction process is set as the peak value Vcospeak of the cosine wave signal.

On the other hand, if FLAG_hosei = 0 in step S51, for example, a sine wave signal is acquired at timings t12 and t16, and A / D conversion is performed (step S52). The voltage value range of the sine wave signal is −1 to +1.
In step S53, the voltage value of the acquired sine wave signal is set to Vsin1, and correction processing is performed on this Vsin1 in the next step S54. In the correction process, the calculation of “Vsin1hosei = Vsin1 × correction coefficient” is performed. The correction coefficient is 1.11 when the amplitude ratio α = 0.9. In step S55, Vsin1hosei obtained by the correction process is set as the peak value Vsinpeak of the sine wave signal.

FIG. 13 is a flowchart showing the voltage acquisition processing operation on the peak side. In step S61, it is determined whether the correction target is a sine wave signal. For this determination, the correction side update flag FLAG_hosei is used. If FLAG_hosei = 0, the voltage value of the cosine wave signal is acquired, and if FLAG_hosei = 1, the voltage value of the sine wave signal is acquired. For example, at the timings of times t1, t5, and t9, the voltage value of the sine wave signal is acquired and A / D conversion is performed (step S65). The voltage value range of the sine wave signal is −1 to +1.
In step S66, the voltage value of the acquired sine wave signal is set to Vsin2, and this Vsin2 is set to the peak value Vsinpeak of the sine wave signal (step S67).

On the other hand, if FLAG_hosei = 0 in step S61, for example, the voltage value of the cosine wave signal is acquired at the timings t13 and t17, and A / D conversion is performed (step S62). The voltage value range of the cosine wave signal is −1 to +1.
In step S63, the voltage value of the acquired cosine wave signal is set to Vcos2, and this Vcos2 is set to the peak value Vcospeak of the cosine wave signal (step S64).

In the next step S68, the absolute value of the peak value Vcospeak of the cosine wave signal in step S64 or S59 is compared with the absolute value of the peak value Vsinpeak of the sine wave signal in step S67 or S55. When the relationship of “| Vcospeak | ≧ | Vsinpeak |” is satisfied, the flag FLAG_hosei is set to “1” (step S69). Otherwise, the flag FLAG_hosei is cleared to “0” (step S70).
In the next step S71, the motor angle is calculated. In this calculation, the motor angle is calculated using the arc tangent from the peak value Vsinpeak of the sine wave signal and the peak value Vcospeak of the cosine wave signal.

  According to the fourth detection operation as described above, both the two signal waveforms output from the resolver can be handled as the output value of the peak position as in the first to third detection operations. Reduction can be suppressed. In addition, by adopting a configuration in which the correction processing is performed on the waveform having the larger amplitude of the cosine wave signal and the sine wave signal, the correction processing is not performed in a small amplitude region with poor resolution, so that the accuracy can be improved.

[Fifth Embodiment]
The fifth embodiment is an example of control when an abnormality occurs in the first to fourth embodiments described above. Abnormality is detected by storing the acquired voltage value and calculated value, and determining that an abnormality has occurred when a sudden change has occurred compared to the previous value. When it is determined that the detected voltage value is abnormal, the angle calculation is performed using the previous values of both the voltage values of the sine wave signal and the cosine wave signal. Alternatively, only the voltage value of one signal determined to be abnormal is used for the previous value, and the other is used for angle calculation using the current value.

Similarly, it is determined that the voltage value calculated for correction is abnormal when a sudden change occurs compared to the previous time. If it is determined that there is an abnormality, the angle is calculated using the acquired value before correction. Alternatively, the previous angle calculation value is used.
Thus, by determining abnormality using the learned previous value, it is possible to suppress the influence when abnormality occurs.

<Application example>
FIG. 14 is a schematic configuration diagram when the rotation angle detection device for a motor according to the embodiment of the present invention is applied to a VCR (Variable Compression Ratio). The engine 11 is equipped with a VCR mechanism 12. The VCR mechanism 12 is driven by a brushless motor (electric actuator) M, and changes the compression ratio of the engine 1 according to the running state. For example, the VCR mechanism 12 rotates the control shaft under the control of the brushless motor M and moves the control point, thereby changing the compression ratio by changing the piston position at the top dead center.

  A resolver (relative angle sensor) 13 for detecting the rotation angle of the motor is attached to the brushless motor M, and a sine wave signal and a cosine wave signal output from the resolver 13 are converted into a VCR controller (VCR-CU) 14. Is input. The engine control unit (ECU) 15 calculates a target value of the control position of the VCR mechanism 12 and transmits it to the VCR controller 14 by CAN (Controller Area Network). The VCR controller 14 controls the brushless motor M with a three-phase current so that the received target value is obtained. The control shaft is rotated by the control of the brushless motor M, and the compression ratio is changed by the change of the piston position at the top dead center.

  The resolver 13 receives an excitation signal from the ECU 15 or the VCR-CU 14 and outputs a sine wave signal and a cosine wave signal corresponding to the rotation angle of the motor M. The voltage values of the sine wave signal and the cosine wave signal are acquired, the rotation angle is calculated by the VCR-CU 14 and supplied to the ECU 15. In this example, the VCR-CU 14 corresponds to the motor control device 2 in FIG. 1 and also functions as the A / D converter 4 and the arithmetic device 5. For example, the VCR-CU 14 calculates the rotation angle of the brushless motor M according to the procedure described in the fourth detection operation, and feedback-controls the VCR mechanism 12 by the VCR-CU 14 based on the rotation angle.

  FIG. 15 is a waveform diagram for explaining the relationship between the rotation angle of the motor and the voltage values of the sine wave signal and the cosine wave signal in the VCR shown in FIG. In a system that controls the compression ratio of an engine, the frequency of use is high when the compression ratio is maintained at a high compression ratio and when the compression ratio is maintained at a low compression ratio. Therefore, it is desirable that the relationship between the rotation angle and the compression ratio is configured so as to avoid a region having a small amplitude from being a region that is frequently used. Therefore, the motor angle in the region where the compression ratio is low and the region where the compression ratio is high is in the range of 45 ° to 315 °.

  That is, the regions A1 and A2 where the motor angle is near 90 ° and 270 ° enter the region B where the voltage value of the cosine wave signal is near 0, and the region A3 where the motor angle is near 180 ° is the voltage value of the sine wave signal. Enters region B near zero. On the other hand, in a region C1 where the compression ratio is low at a motor angle of about 45 °, both the voltage values of the sine wave signal and cosine wave signal have relatively large amplitudes, and a region C2 where the compression ratio is high at a motor angle of about 315 °. Although the voltage value of the sine wave signal is positive and the voltage value of the cosine wave signal is negative, the amplitude is still relatively large. Therefore, the areas C1 and C2 having a large amplitude are frequently used areas.

  As described above, in the present invention, the amplitude of the excitation signal is maximized for either the sine wave or the cosine wave even when the A / D conversion cannot be performed at the same timing for the sine wave signal and the cosine wave signal. Obtain the voltage of the point output signal, and obtain the voltage value of the sine wave signal and cosine wave signal at substantially the same timing by performing a simple process by acquiring the voltage by shifting the timing and performing the correction. Can do. In addition, since the voltage value of the output signal can be acquired at the point where the amplitude is maximum, the resolution is increased, the voltage value of the output signal can be acquired more accurately, and the motor rotation angle can be accurately detected. .

The present invention is not limited to the first to fifth embodiments and application examples described above, and can be applied to any apparatus that performs angle detection using a resolver in motor control.
(Modification 1)
In each of the above embodiments, the voltage value of the sine wave signal is detected at a timing before a predetermined time with respect to the peak position of the excitation signal, but the voltage value of the sine wave signal at a timing after the predetermined time. Can also be detected.

(Modification 2)
Further, in each of the above embodiments, the rotation angle detection device 1 includes the resolver 3, the A / D converter 4, and the arithmetic device 5, and the sine wave signal and the cosine wave signal output from the resolver 3 are converted into the A / D converter 4. A case has been described in which the rotation angle of the motor is detected by processing in the arithmetic unit 5. However, as shown in the application example, part or all of the functions of the A / D converter 4 and the arithmetic unit 5 are realized by using a processing device such as a motor control device or a microcomputer provided in the ECU. You can also.

  DESCRIPTION OF SYMBOLS 1 ... Rotation angle detection apparatus, 2 ... Motor control apparatus, 3, 13 ... Resolver, 4 ... A / D converter, 5 ... Arithmetic unit, M ... Motor (electric motor)

Claims (4)

A rotation angle detection device that includes a resolver that outputs first and second signals according to the rotation angle of the electric motor, and detects the rotation angle of the electric motor,
Based on the voltage value at the peak position of the amplitude in the first signal output from the resolver and the voltage value at a timing shifted from the peak position of the amplitude in the second signal by a predetermined time, the rotation angle of the electric motor is determined. An apparatus for detecting a rotation angle of a motor.   2. The motor rotation angle detection device according to claim 1, wherein the voltage value at a timing shifted by a predetermined time is corrected by an amplitude ratio with an amplitude peak position in an excitation signal input to the resolver. .   3. The motor rotation angle detection device according to claim 1, wherein the timing shifted by a predetermined time is a timing before a predetermined time with respect to an amplitude peak position in the second signal.   The voltage value of the 1st, 2nd signal output from the said resolver is A / D-converted with one A / D converter, The one of Claim 1 thru | or 3 characterized by the above-mentioned. Motor rotation angle detection device.