JP2001343253A - Method of detecting abnormality of resolver - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method of detecting the abnormality of a resolver which judges the resolver to be in the abnormality in a wide rotation angle range by a simple computation. SOLUTION: If the abnormality such as short circuit occurs in an output coil of a resolver, a sampling value of the output signal from the abnormality- occurred coil about the rotation angle X is offset. This value is obtained as a value across the abnormality occurred coil at a timing at which the other coil to the abnormality-occurred coil provides a maximum or minimum output. If, thus, the offset value is out of a specified range near 0, the resolver can be judged to be in failure.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an improvement in a method for detecting an abnormality of a resolver.

[0002]

2. Description of the Related Art Conventionally, a resolver has been used to detect a rotor rotation angle of a rotating machine such as an electric motor. The rotation angle detected by the resolver is used for controlling the current of the rotating machine. Therefore, when an abnormality such as a short-circuit of the coil occurs in the resolver, the control of the rotating machine cannot be performed normally, so that it is necessary to detect the abnormality of the resolver promptly.

For example, Japanese Unexamined Patent Application Publication No. 64-80817 discloses an abnormality detection device for such a resolver (sine wave encoder). In this conventional example, the sum of the squares of the sine signal and the cos signal output from the resolver is obtained, and when this value deviates from 1 by a predetermined value or more, it is determined that the resolver is abnormal.

[0004]

However, in the above-described conventional resolver abnormality detection device, the threshold value is set in consideration of signal variations based on the accuracy error of the abnormality detection device. As the distance increases, there has been a problem that the rotation angle range of the rotating device that cannot detect the abnormality of the resolver becomes large.

Another problem is that the calculation of the sum of squares of the sin signal and the cos signal is complicated.

SUMMARY OF THE INVENTION The present invention has been made in view of the above-mentioned conventional problems, and an object of the present invention is to provide a resolver abnormality detection method which can determine a resolver abnormality in a wide rotation angle region by a simple calculation. .

[0007]

In order to achieve the above object, the present invention relates to a method for detecting an abnormality of a resolver, which samples each resolver output signal in synchronization with the maximum value or the minimum value of the excitation signal of the resolver. When the offset of the sampling value is out of a predetermined range near 0, the resolver is determined to be abnormal.

Further, in the above-described method for detecting an abnormality of a resolver, the offset is the other value when one of the sampling values is the maximum value or the minimum value.

Further, in the above-described method for detecting an abnormality of a resolver, the offset is an average value of a maximum value and a minimum value of each sampling value.

A method for detecting an abnormality of a resolver,
Each resolver output signal is sampled after a predetermined time from the time when the excitation signal of the resolver crosses zero voltage, and the predetermined time is gradually changed to find a time when each resolver output signal becomes zero, and the excitation signal crosses zero voltage. When a difference between the time when the resolver output signal becomes zero and the time when each resolver output signal becomes 0 is out of a predetermined range, the resolver is determined to be abnormal.

[0011]

Embodiments of the present invention (hereinafter referred to as embodiments) will be described below with reference to the drawings.

FIG. 1 shows an example of a configuration for implementing a method for detecting an abnormality of a resolver according to the present invention. In FIG. 1, a resolver 10 for detecting a rotation angle of a rotating machine includes an excitation coil 12 to which an excitation signal (REF signal) is applied, a sin coil 14 for generating an output signal based on the excitation signal, and a c.
The os coil 16 is provided.

The sin coil 14 and the cos coil 16 generate a sin coil signal and a cos coil signal, respectively, as output signals. These two output signals are sent to the CPU 22 via the differential amplifiers 18 and 20, respectively.
And converted to digital by an analog-to-digital converter (ADC).

The REF signal is input via a differential amplifier 24 to a maximum value search circuit 26 composed of a comparator and the like. The maximum value search circuit 26 outputs a synchronization signal at the timing of the maximum value or the minimum value of the REF signal,
This synchronization signal is input to the CPU 22.

In the CPU 22, while the rotating machine whose rotation angle is to be detected by the resolver 10 is rotating, the synchronization signal output from the maximum value search circuit 26, that is, the output of the resolver 10 at the timing when the REF signal reaches the maximum value or the minimum value. Sin coil signals and cos coil signals, which are signals, are sampled. This situation will be described with reference to FIGS.

FIG. 2 shows a sine coil signal generated by the sine coil 14 and the cos coil 16 when a REF signal is applied to the exciting coil 12 in a state where the rotating machine whose rotation angle is to be detected by the resolver 10 is stopped. The cos coil signal is shown. As can be seen from FIG. 2, the sin coil signal and the cos coil signal have the same phase as the REF signal.

On the other hand, FIG.
s when an abnormality such as a short circuit occurs in the os coil 16
The in coil signal and the cos coil signal are shown. Also in FIG. 3, the rotating machine is in a stopped state. As shown in FIG. 3, when a short circuit occurs in the cos coil 16, the phase of the cos coil signal is shifted from the phase of the REF signal. On the other hand, the sine coil signal output from the sine coil 14 in which no short circuit occurs has the same phase as the REF signal as in FIG.

Such sin coil signal and cos
If the coil signal is sampled at a timing when the REF signal reaches a maximum value while rotating the rotating machine, in the case of FIG. 2 where no abnormality has occurred, a waveform as shown in FIG. 4 is obtained. The result of sampling the sin coil signal is the SIN signal, and the result of sampling the cos coil signal is the COS signal. When the rotating machine rotates, the amplitudes of the sin coil signal and the cos coil signal change in phase with the REF signal in accordance with the rotation angle X, and the sampling results are as shown in FIG. When the rotation angle X is taken on the horizontal axis, the SIN signal and the COS signal are
The sine wave signal is out of phase by 90 °.

On the other hand, in the case of FIG. 3 in which an abnormality such as a short circuit has occurred in the cos coil 16, the degree of the phase shift between the REF signal and the cos coil signal changes with the rotation of the rotating machine. Accordingly, depending on the rotation angle X, the phase may be the same as that of the REF signal, or a phase shift may occur as shown in FIG. Such a co
When the s-coil signal and the sin-coil signal in which no abnormality has occurred are sampled at a timing when the REF signal reaches a maximum value while rotating the rotating machine, a waveform as shown in FIG. 5 is obtained. 4, when the rotation angle X is plotted on the horizontal axis, the SIN signal and the COS signal have a phase of 90 degrees.
Although the sine wave signal is shifted by °, the COS signal is shifted upward in the figure. Accordingly, the maximum value (point P) and the minimum value (point R) of the COS signal are also shifted upward from the normal values. For this reason, an offset is generated by an amount corresponding to the shift. This offset depends on the abnormal state generated in the COS coil 16.
It may occur in the downward direction in the figure.

In FIGS. 2 and 3, the sin coil signal and the cos signal are output at the timing when the REF signal has the maximum value.
Although the coil signal is sampled, a similar result can be obtained by sampling at a timing when the REF signal has a minimum value.

As described above, the phase relationship between the SIN signal and the COS signal is constant in both FIGS. 4 and 5, and the relationship between the sin curve and the cos curve is given. The other takes an intermediate value between the maximum value and the minimum value. For this reason, in FIG. 4, at the timing of the maximum value or the minimum value of the SIN signal, C
The value of the OS signal becomes 0, and conversely, the SIN signal becomes 0 at the timing of the maximum value or the minimum value of the COS signal.
In contrast, in FIG. 5, the COS signal takes an offset value at the timing of the maximum value or the minimum value of the SIN signal.

As described above, the sine coil signal and the cos coil signal which are two output signals of the resolver 10 are sampled in synchronization with the maximum value or the minimum value of the REF signal of the resolver 10, and the offset of the sampled value is close to zero. When it is out of the predetermined range, it can be determined that the resolver 10 is abnormal. This is because, as shown in FIG. 5, when a short circuit occurs in the cos coil 16, for example, the COS signal is not between the normal maximum value and the minimum value, but an offset occurs.

Such an offset is, as shown in FIG. 5, one sampling value of the output signal of the resolver 10, in FIG. 5, the other sampling value when the value of the SIN signal is the maximum value or the minimum value, that is, It can be obtained as the value of the COS signal. All such offset detection is performed by the CPU 22. In this case, the sin coil signal and cos
If the coil signal is divided by the REF signal, the influence of common-mode noise included in the sine coil signal, the cos coil signal, and the REF signal can be eliminated, and the possibility of a maximum value capture error can be reduced.

The calculation of the offset value performed by the CPU 22 includes the SIN signal or C
A method of calculating an average value of the maximum value and the minimum value of the OS signal may be used. For example, as shown in FIG. 5, when an offset occurs in the COS signal, by taking the average of the values of the maximum point P and the minimum point R deviated from the normal maximum value and minimum value, This is the offset value.

By detecting the offset of the SIN signal or the COS signal which is the sampling value by the above method and detecting the abnormality of the resolver, the half cycle of the SIN signal or the COS signal, that is, the rotation angle X is 18
Abnormality of the resolver 10 can be determined within the range of 0 °.

The sin coil signal and the cos coil signal are also input to the angle converter 27. Angle converter 27
Then, the rotation angle X measured by the resolver 10 is calculated from the sine coil signal and the cos coil signal by analog processing, and a north marker (NM) signal is output at timings when the rotation angle X is 0 ° and 180 °. . This NM is C
The input to the PU 22 is used for the operation of the CPU 22 shown in FIGS.

FIG. 6 is a flowchart of the steps of the method for detecting an abnormality of the resolver 10 described above. In FIG. 6, first, it is confirmed whether or not the first North Marker (NM) has been input (S1). The north marker is a signal input to the CPU 22 at the timing when the rotation angle X measured by the resolver 10 is 0 ° and 180 ° as described above. Therefore, between each north marker, the SIN signal and the COS signal, which are the sampling values, always take the maximum value or the minimum value once.

Next, in synchronization with the maximum value or the minimum value of the REF signal, the CPU 22 samples the sin coil signal and the cos coil signal (S2), and stores the sampled values in a memory (S3).

It is determined whether the sampling value is the maximum value among the input sampling values (S4). If the sampling value is the maximum value, the stored sampling value is determined as the maximum value. Replace it and update the maximum value. Also, a sampling value of a coil signal different from the coil signal on the side where the maximum value is obtained is stored as a median value (S
5).

If the sampling value is not the maximum value in S4, the step of S5 is omitted. The above operations S2 to S5 are repeated until the second input of the north marker (S6).

As described above, the designed median value is subtracted from the median value of one coil signal, for example, the COS signal on the other side when the SIN signal takes the maximum value in the example shown in FIG. An offset value is obtained (S7). It is determined whether the offset value is within a predetermined range, for example, near 0 (S8). If the offset value is within the predetermined range, it is determined that the offset value is normal (S9). If the offset value is out of the range, the resolver is abnormal. It is determined that there is (S10).

Thus, the resolver abnormality detecting step is completed.

FIG. 7 shows a flowchart of a process for obtaining the maximum value and the minimum value of the REF signal for the purpose of generating a synchronization signal for causing the CPU 22 to take the timing of sampling. The operation shown in this flowchart is performed by the maximum value search circuit 26 in FIG.

First, an initial timer value is set (S10).
1) Then, a predetermined change time is added to the next set timer value, and the current timer value is set (S102).

The REF signal is sampled at the timing when the timer value expires (S10).
3). The value of the sampled REF signal is stored in the memory (S104).

Next, it is determined whether or not the sampling value stored in the memory is the largest among the sampling values stored conventionally (S105). If the maximum value is reached, the maximum value is updated (S10
6).

If the stored sampling value is not the maximum value, it is determined whether the sampling value is the minimum value (S107). If it is the minimum value, the minimum value is updated (S108). If the sampling value is not the minimum value, step S108 is omitted.

Next, it is determined whether or not to continue the search for the maximum value or the minimum value described above.
The steps from 102 are repeated (S109).

Through the above steps, the maximum and minimum values of the REF signal can be obtained.

When the search for the maximum value or the minimum value is not to be continued, the amplitude is obtained by taking 1/2 of the difference between the maximum value and the minimum value. Further, an offset is obtained by taking the difference between the value obtained by halving the sum of the maximum value and the minimum value and the design median value (S11).
0).

It is determined whether the amplitude and offset obtained as described above are normal (S111), and a normal (S112) or abnormal (S113) signal is output, respectively.

Through the above steps, the timing of the maximum or minimum value of the REF signal can be known,
It is also possible to determine whether the REF signal itself is normal or abnormal in the steps after S110.

FIG. 8 shows a flowchart of a process for obtaining an offset from the average value of the maximum value and the minimum value of the SIN signal or COS signal which is a sampling value.

In FIG. 8, first, the first north marker is input as in FIG. 6 (S201).

Next, the CPU 22 samples the sin coil signal and the cos coil signal in synchronization with the synchronizing signal indicating the maximum value or the minimum value of the REF signal (S202). The sampled value is stored in the memory (S203).

Next, it is determined whether or not the sampling value stored in the memory is the maximum value among the sampling values stored so far (S204). If it is the maximum value, the maximum value is updated (S205). The determination of the maximum value is performed based on the sampling values of the sin coil signal and the cos coil signal, that is, both the SIN signal and the COS signal.

If the sampling value is not the maximum value in S204, it is determined whether or not the sampling value is the minimum value among the values stored so far (S2).
06). If it is the minimum value, the minimum value is updated (S207). If it is not the minimum value, the step of S207 is omitted.

The operations from S202 to S207 are two
This is repeated until the input of the north marker for the second time (S20)
8).

The SIN signal and C
An average value is calculated from the maximum value and the minimum value of the OS signal (S209). Further, an offset is calculated from the difference between the average value calculated in S209 and the design median value (S209).
210). Next, it is determined whether the offsets of the SIN signal and the COS signal thus obtained are normal (S211). If normal, a normal signal is output (S21
2) If abnormal, that is, if it is out of the predetermined range near 0, an abnormal signal is output (S213).

FIGS. 9A and 9B show the sin coil 1
4 or cos coil 16 has an abnormality such as short circuit,
A modification of the present invention relating to detection of a phase difference when a phase difference occurs between a REF signal and a sin coil signal or a cos coil signal is shown. In FIG. 9A, the CPU 22
The REF signal, the sine coil signal, and the c
The os coil signal is input, and a zero cross trigger signal indicating the timing at which the REF signal crosses zero voltage is input from the comparator 28 as an interrupt signal (INT signal). In the CPU 22, the sin coil signal and the cos
The coil signal is sampled.

When no abnormality such as a short circuit has occurred in the sin coil 14 and the cos coil 16, the phases of the REF signal, the sin coil signal, and the cos coil signal are all the same as shown in FIG. On the contrary,
For example, when an abnormality occurs in the cos coil 16, the phase of the cos coil signal is shifted from the phase of the REF signal as shown in FIG. Therefore, FIG.
The interrupt signal is output from the comparator 28
When the sampling of the cos coil signal is performed at the timing input from the oscilloscope, that is, at the timing when the REF signal becomes 0, the phase shift can be detected since the value is not 0.

FIG. 9B is an explanatory diagram for detecting the degree of the phase shift. In FIG. 9B, an interrupt signal (INT signal)
The timing at which the REF signal becomes 0 is shown at the timing indicated by に. The sampling of the sine coil signal and the cos coil signal is performed at a timing when a predetermined time t has elapsed from the point when the REF signal becomes 0. The value of the time t that defines this sampling timing is shown in FIG.
As shown by the ▼ in (b), when gradually shortening,
It is possible to detect a point in time at which the value of a cos coil signal generated from a coil in which an abnormality has occurred at a certain point, for example, the cos coil 16 becomes zero. If no abnormality occurs, the cos coil signal should be 0 at the timing when the REF signal becomes 0 as described above. The phase angle difference between the signals can be obtained. Accordingly, if the difference between the timing of the 0 cross of the REF signal and the timing at which the sin coil signal or the cos coil signal becomes 0 is out of the predetermined range, it means that the phase angle difference has increased. It can be determined that the resolver 10 is abnormal.

[0053]

As described above, according to the present invention,
Since abnormality of the resolver is detected based on the offset of the sampling signal, abnormality of the resolver can be determined within a rotation angle of 180 °.

The offset value is obtained from the other value when one of the sampling values is the maximum value or the minimum value, or from the average value of the maximum value and the minimum value. Calculation can be easily performed.

Further, by detecting the abnormality of the resolver from the difference between the timing of the excitation signal of 0 and the timing of the resolver output signal of 0, the abnormality of the resolver can be detected with a simple configuration.

[Brief description of the drawings]

FIG. 1 is a diagram showing a configuration example for implementing a resolver abnormality detection method according to the present invention.

FIG. 2 is a diagram illustrating an example of a REF signal, a sine coil signal, and a cos coil signal in a normal state.

FIG. 3 is a diagram illustrating an example of a REF signal, a sine coil signal, and a cos coil signal when an abnormality occurs in a cos coil.

FIG. 4 is a diagram showing a change in a value obtained by sampling a value of a sine coil signal and a value of a cos coil signal at a timing of a maximum value or a minimum value of the REF signal shown in FIG. 2 with respect to a rotation angle.

FIG. 5 is a diagram illustrating a change in a value obtained by sampling a value of a sine coil signal and a value of a cos coil signal at a timing of a maximum value or a minimum value of the REF signal illustrated in FIG. 3 with respect to a rotation angle;

FIG. 6 is a flowchart of a method of obtaining an offset in CPU 22 shown in FIG. 1;

FIG. 7 is a flowchart for obtaining a maximum value and a minimum value of a REF signal.

FIG. 8 is a flowchart of a method for obtaining an offset in CPU 22 shown in FIG. 1;

FIG. 9 is a diagram showing a modified example of the resolver abnormality detection method.

[Explanation of symbols]

10 resolver, 12 excitation coil, 14 sin coil, 16 cos coil, 18, 20 differential amplifier, 2
2 CPU, 26 maximum value search circuit, 27 angle converter, 28 comparator.

Claims (4)

[Claims] An output signal of each resolver is sampled in synchronization with a maximum value or a minimum value of an excitation signal of the resolver. When an offset of the sampled value is out of a predetermined range near 0, it is determined that the resolver is abnormal. A method for detecting an abnormality of a resolver, characterized by determining. 2. The resolver abnormality detection method according to claim 1, wherein said offset is the other value when one of the sampled values is a maximum value or a minimum value. Anomaly detection method. 3. The resolver abnormality detection method according to claim 1, wherein said offset is an average value of a maximum value and a minimum value of each sampling value. 4. A method of sampling each resolver output signal a predetermined time after the time when the excitation signal of the resolver crosses zero voltage, and gradually changing the predetermined time to obtain a time when each resolver output signal becomes 0, An abnormality detection method for a resolver, comprising determining that the resolver is abnormal when a difference between a time when the excitation signal crosses zero voltage and a time when each resolver output signal becomes 0 is out of a predetermined range.