JP2007139502A - Angle detector and device of calculating rotational coordinate - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To shorten the time required for processing by a simple structure, in regard to an angle detector using a resolver and a device of calculating a rotational coordinate. <P>SOLUTION: This angle detector is equipped with an angle calculation means for calculating a rotation angle of the resolver. The calculation means is characterized by comprising: a count means for counting the number of pulses which are the final output of a phase synchronization means based on an angle zone display signal outputted by a zone identification means and on a final output of the synchronization means; a memory means for therein storing a count value and angle information associated therewith; and an extraction means for extracting the angle information determined by the count value. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an angle detection device that detects a rotation angle of a rotation device in combination with a resolver, and a rotation coordinate calculation device that calculates rotation coordinates, and in particular, to the number of pulses of a pulse signal obtained by multiplying the frequency of a resolver modulation signal. The present invention relates to an angle detection apparatus that performs angle detection based on the rotation coordinate calculation apparatus that performs calculation of rotation coordinates.

As a prior art (first prior art) of an angle detection device that detects a rotation angle, an angle detection device that operates in combination with a one-phase excitation two-phase output type resolver is known. There exists a thing shown by patent document 1 as this conventional angle detection apparatus.
According to the first prior art, a brushless motor control device including an angle detection device composed of a resolver and an electronic control unit, which inputs a sine wave reference signal as an excitation signal to the primary winding of the resolver. The first output signal (COS signal) and the second output signal (SIN signal) obtained by amplitude-modulating the reference signal according to the rotation angle θ are input to the electronic control unit.
Next, after multiplying each output signal by the reference signal, each of the output signals is passed through a low-pass filter to remove the harmonic component. Further, the SIN information component of the signal after passing through the low-pass filter is divided by the COS information component to obtain a TAN (tangent) information component. Based on this, the rotation angle θ is calculated by referring to the tan θ → θ conversion table. is doing.

Another conventional technique (second conventional technique) is composed of a semiconductor integrated circuit called an excitation circuit, an input interface circuit, and an R / D converter (resolver / digital converter). There is an angle detection device combined with a resolver.
As in the prior art described in Patent Document 1, the excitation signal output from the excitation circuit is input to the primary winding of the resolver, and the first output signal (COS signal) obtained by modulating the excitation signal according to the rotation angle θ The second output signal (SIN signal) is transmitted to the R / D converter via the input interface circuit.

  In the R / D converter, for example, the sine value sinφ and cosine value cosφ of the estimated rotation angle φ are multiplied by the first output signal (COS signal) and the second output signal (SIN signal), respectively. The rotation angle θ is calculated in hardware by a technique such as a tracking method in which the rotation angle estimated value φ is converged so that the deviation becomes zero (that is, the rotation angle estimated value φ and the rotation angle actual value θ are equal).

  Further, as a prior art (third prior art) of the rotational coordinate calculation device, a sine value sinθ and a cosine value cosθ are calculated by referring to a θ → sinθ conversion table based on the rotation angle θ calculated from the angle detection device, Some of them use these to calculate rotational coordinates around the origin on the two-dimensional plane shown in Equation 1.

  However, (ux, uy): position before coordinate rotation, (jx, jy): position after coordinate rotation.

Further, as another prior art (fourth prior art) of a rotational coordinate computing device, coordinate conversion between three-phase coordinates and rotational orthogonal coordinates (d, q-axis coordinates) used for vector control computation of a known electric motor is performed. There is something.
The conversion from the three-phase coordinates to the rotation orthogonal coordinates takes the form shown in Equation 2. The trigonometric function values (√2) · sinθ, (√2) · sin (θ + π / 3), (√2) · sin (θ + π / 2), (√ with reference to the θ → (√2) · sinθ conversion table 2) · sin (θ + 5π / 6) is calculated, and the rotational coordinate calculation (coordinate conversion calculation) is performed according to Equation 2.

  However, ju: three-phase coordinate U-phase component, jv: three-phase coordinate V-phase component, jd: rotational orthogonal coordinate d-axis component, jq: rotational orthogonal coordinate q-axis component.

JP 2004-343983 A

  As described above, the angle detection device described in Patent Document 1 requires division of SIN information / COS information in the process of calculating the rotation angle θ. Here, when division is implemented by software processing using a microprocessor together with reference to the tan θ → θ conversion table, which is a mapping of nonlinear characteristics, the time required for the processing becomes long. An increase in processing time leads to an increase in the total calculation amount per unit time in the discrete time control performed by the microprocessor, that is, the load factor. If the load factor is high, the computations to be processed per unit time will not be able to be processed and the stable operation of the processing will be impaired, so it can be replaced with a microprocessor that can perform computations at higher speeds, or discrete time control It is necessary to take measures such as extending the unit time. Replacement with a high-speed arithmetic microprocessor has a problem that the cost of the apparatus increases, and extension of unit time of discrete time control has a problem that it leads to performance deterioration such as control responsiveness and accuracy deterioration.

  In the angle detection device according to the second prior art, since the processing of the microprocessor necessary for calculating the rotation angle θ is replaced by hardware processing by the R / D converter by using the R / D converter, Although the problem caused by the increase is solved, there is a problem that the cost is greatly increased by using the R / D converter.

  In the third and fourth conventional rotational coordinate arithmetic devices, the rotation angle θ is calculated, and then the sine value sinθ and cosine value cosθ are calculated with reference to the θ → sinθ conversion table. Therefore, it is necessary to calculate the rotation angle θ with higher accuracy because it is necessary to determine the resolution of the rotation angle θ in accordance with the rotation angle, or rotate to match the configuration of the table elements to apply the θ → sin θ conversion table. There is a problem in that redundant processing is required, for example, it is necessary to perform preprocessing for converting the numerical data representation of the angle θ.

  The present invention has been made to solve the above-mentioned problems. Even when combined with a discrete time control device, the control response and accuracy can be maintained and improved at a low cost while shortening the processing time. An object of the present invention is to provide an angle detection device that can be realized.

  In addition, the present invention is a rotational coordinate computing device that can realize processing at a low cost and in a short time required for computing rotational coordinates based on a trigonometric function value of the rotational angle θ in combination with a discrete time control device. The purpose is to provide.

  An angle detection apparatus according to the present invention is an angle detection apparatus for detecting a rotation angle of a resolver that receives an excitation signal of a periodic function and outputs an amplitude modulation signal or a phase modulation signal using the excitation signal as a carrier wave. Demodulating means for demodulating the modulation signal, section identifying means for identifying an angle section based on the demodulated signal demodulated by the demodulating means, and outputting an angle section display signal representing the angle section according to the identification result, section identifying means Based on the identification result, the pulse signal generating means for generating the pulse signal, and the pulse signal generated from the pulse generating means as a reference signal, and the reference signal and the comparison signal obtained by multiplying the final output are compared in phase. Based on the phase synchronization means for obtaining a pulsed final output, the angle section display signal output by the section identification means, and the final output of the phase synchronization means, Angle calculating means for calculating the rotation angle of the solver, and the angle calculating means is based on the angle section display signal output by the section identifying means and the final output of the phase synchronizing means, and the number of pulses of the final output of the phase synchronizing means. And a memory means for storing angle information associated with the count value, and an extracting means for extracting angle information determined by the count value.

  According to the present invention, since a division operation in the process of calculating the rotation angle θ is unnecessary, even when combined with a discrete time control device while reducing the processing time, control responsiveness and It is possible to provide an angle detection device that maintains and improves accuracy. Further, since the R / D converter is not used, the cost required for applying the R / D converter is unnecessary, and an inexpensive angle detection device can be provided.

  Further, according to the present invention, the number of pulses per one rotation angle of the frequency-multiplied pulse signal is equal to the total number of rotation angle expression data per one rotation angle determined based on the required data resolution of the rotation angle θ. If it is determined, it can be regarded as the number of measured pulses starting from the rotation angle reference (= 0 degree) and the amount of change from the reference (= 0 degree), and the rotation angle at that time can be correlated. Since it becomes simple, redundant data conversion processing is not required in calculating the rotation angle based on the number of measured pulses, and an angle detection device that further shortens the time required for calculation processing can be provided.

  Further, according to the present invention, the trigonometric function value of the rotation angle θ used for the coordinate rotation calculation is not a rotation angle represented by data, but a pulse starting from the rotation angle reference (= 0 degree) of the frequency-multiplied pulse signal. A rotation coordinate calculation device that shortens the time required for calculation processing is provided by calculating by associating an offset amount with a reference (= 0 degree) of a table storing trigonometric function values over one rotation angle period for the measured number of rotations can do. Further, the number of pulses per rotation angle period of the frequency-multiplied pulse signal is set equal to the total number of rotation information data including trigonometric function values per rotation angle period determined based on the required data resolution of the coordinate conversion calculation. If this is the case, the number of pulses measured with the rotation angle reference (= 0 degree) as the starting point and the rotation information data at that time can be regarded as the amount of change from the reference (= 0 degree) is simple. Therefore, a redundant data conversion process is not required in the calculation of rotation information data based on the measured number of pulses, and a rotation coordinate calculation apparatus with a reduced calculation process time can be provided.

  In addition, according to the present invention, since it is possible to reduce the time required for calculation processing relating to calculation of rotation information data including trigonometric function values, rotation coordinates with reduced calculation processing time for coordinate rotation in an integer L-dimensional space including a two-dimensional plane An arithmetic device can be provided.

  Further, according to the present invention, since the time required for the calculation processing related to the calculation of rotation information data including trigonometric function values can be shortened, coordinate conversion from three-phase AC coordinates to rotation orthogonal coordinates, or rotation orthogonal coordinates to three-phase It is possible to provide a rotating coordinate calculation device that shortens the calculation processing time for coordinate conversion to AC coordinates.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing the overall configuration of the angle detection apparatus according to Embodiment 1 of the present invention.
In FIG. 1, a resolver 2 is connected to the angle detection device 1.
The angle detection device 1 includes an excitation circuit 3, an input interface circuit 4, a demodulation unit 9, a pulse signal generator 10, a phase synchronization circuit 11, a section identification unit 12, a microprocessor 13, and an angle calculation unit 14.

The resolver 2 includes a primary winding (excitation) 21, a secondary winding (COS) 22 arranged as a phase difference of 90 degrees, a secondary winding (SIN) 23, and a rotor 24.
The angle detection device 1 and the resolver 2 are connected by signal transmission lines R1, R2, S1, S3, S2, and S4.

The operation of the angle detection apparatus according to Embodiment 1 of the present invention is as follows.
First, the excitation circuit 3 generates an excitation signal ref_sig having a periodic function as a reference signal and inputs it to the primary winding (excitation) 21 through the signal transmission lines R1 and R2. As a result, an induced voltage obtained by modulating the amplitude of the reference signal in a cosine function according to the rotation angle θ of the rotor 24 is induced in the secondary winding (COS) 22, and the induced voltage modulated in a sine function is It is induced in the next winding (SIN) 23. The values of the respective induced voltages are transmitted through the signal transmission lines S1 and S3 as the cosine function modulation signal mod_cos, and are input through the signal transmission lines S2 and S4 as the sine function modulation signal mod_sin. 4 is transmitted.

FIG. 2 shows a waveform example of the rotation angle θ, the excitation signal ref_sig, the cosine function modulation signal mod_cos, and the sine function modulation signal mod_sin with respect to the time axis as a common horizontal axis.
FIG. 2A shows a state in which the rotation angle θ of the resolver 2 with respect to the time axis increases monotonously.
FIG. 2B shows a change in the excitation signal ref_sig.
FIG. 2C shows a change in the cosine function modulation signal mod_cos.
FIG. 2D shows changes in the sinusoidal modulation signal mod_sin.

In the input interface circuit 4 in FIG. 1, the cosine function modulation signal mod_cos is amplified by the amplification circuit 41 and shaped in waveform, and the sine function modulation signal mod_sin is amplified by the amplification circuit 42 and shaped in waveform. .
Based on the excitation signal ref_sig and the outputs of the amplifier circuits 41 and 42, the demodulator 9 that demodulates the modulated signal of the resolver demodulates the cosine function demodulated signal dem_cos and the sine function demodulated signal dem_sin based on the output of the amplifier circuits 41 and 42, respectively. The

FIG. 3 shows a cosine function-like modulation signal mod_cos, a cosine function-like demodulation signal dem_cos, a sine function-like modulation signal mod_sin, and a sine function-like demodulation signal when the rotation angle θ changes as shown in FIG. The example of a waveform of dem_sin is shown.
FIG. 3A illustrates a change in the cosine function modulation signal when the excitation signal ref_sig is zero or a positive value.
FIG. 3B illustrates a change in the cosine function modulation signal when the excitation signal ref_sig is a negative value.
FIG. 3C shows a change in the cosine function demodulated signal obtained by synthesizing the waveforms of FIG. 3A and FIG.
FIG. 3D illustrates a change in the sinusoidal modulation signal when the excitation signal ref_sig is zero or a positive value.
FIG. 3E illustrates a change in the sinusoidal modulation signal when the excitation signal ref_sig is a negative value.
FIG. 3 (f) shows the change of the sinusoidal demodulated signal obtained by synthesizing the waveforms of FIG. 3 (d) and FIG. 3 (e).

Here, the cosine function-like demodulated signal dem_cos and the sine function-like demodulated signal dem_sin can be divided into four sections from 0 to 2π in the change range of the rotation angle θ by the combination of positive and negative.
FIG. 4 shows the division characteristics of the cosine function demodulated signal dem_cos and the sine function demodulated signal dem_sin.
That is, as shown in FIG. 4, when dem_cos is zero or a positive value and dem_sin is zero or a positive value, the rotation angle θ exists in the interval from 0 to π / 2.
Further, when dem_cos is a negative value and dem_sin is zero or a positive value, the rotation angle θ exists in the interval from π / 2 to π.
Further, when dem_cos is a negative value and dem_sin is a negative value, the rotation angle θ exists in the interval from π to 3π / 2.
Further, when dem_cos is zero or a positive value and dem_sin is a negative value, the rotation angle θ exists in the interval from 3π / 2 to 2π.

The section identifying means 12 identifies an angle section based on the demodulated signal demodulated by the demodulating means 9 and outputs an angle section display signal representing the angle section according to the identification result. Specifically, the section identifying means 12 generates angle section identifying signals sgn_cos and sgn_sin for identifying an angle section by binary logic of “H” and “L” based on the above-described segmentation characteristics.
FIG. 5 shows the relationship between the output logic of the angle section identification signal and the sign of the cosine function demodulated signal dem_cos and the sign of the sine function demodulated signal dem_sin, respectively.
When dem_cos is zero or a positive value, sgn_cos outputs 'H' logic, and when dem_cos is a negative value, sgn_cos outputs 'L' logic.
When dem_sin is zero or positive, sgn_sin outputs “H” logic, and when dem_sin is negative, sgn_sin outputs “L” logic.

Subsequently, the pulse signal generator 10 which is a pulse signal generation unit generates a pulse signal based on the identification result of the section identification unit 12. That is, the pulse signal generator 10 synthesizes the binary logic of the angle section identification signals sgn_cos and sgn_sin by exclusive OR, and generates and outputs one pulse signal pls_cors.
FIG. 6 shows a waveform example of the rotation angle θ, the angle section identification signals sgn_cos and sgn_sin, and the pulse signal pls_cors with respect to the time axis that is a common horizontal axis.
FIG. 6A shows a change in which the rotation angle θ of the resolver 2 with respect to the time axis increases monotonously.
FIG. 6B shows a change in the angle section identification signal sgn_cos.
FIG. 6C shows a change in the angle section identification signal sgn_sin.
FIG. 6D shows changes in the pulse signal pls_cors.
As shown in these figures, sgn_cos has a logic of “H” when the rotation angle θ is in the range of 0 to π, and “L” otherwise.
Further, sgn_sin is “H” in the range of π / 2 to 3π / 2, and is “L” in other cases.
The pulse signal pls_cors is synthesized as an exclusive OR of sgn_cos and sgn_sin, and is “H” when the rotation angle θ is in the range of 0 to π / 2 and π to 3π / 2, and is otherwise “L”. The logic of '.

  The pulse signal pls_cors is input to the phase synchronization circuit 11 and the frequency is multiplied by N, and then output as a frequency multiplied pulse signal pls_fine as shown in FIG.

FIG. 7 is a block diagram showing a detailed configuration of the phase synchronization circuit 11 of the angle detection apparatus according to Embodiment 1 of the present invention.
As shown in FIG. 7, the phase synchronization circuit 11 includes a phase comparator 110, a loop filter 111, a voltage controlled oscillator 112, and a 1 / N divided period 113. This is a known configuration in which the phase synchronization circuit is applied as a frequency multiplication circuit.

The phase synchronization circuit 11 serving as phase synchronization means uses the pulse signal generated from the pulse signal generator 10 as a reference signal, and compares the phase of the reference signal with a comparison signal obtained by multiplying the final output of the phase synchronization circuit 11. As a result, a pulse-like final output is obtained.
Specifically, the operation of the phase synchronization circuit 11 is as follows.
First, the pulse signal pls_cors output from the pulse signal generator 10 is input to the phase comparator 110 in the phase synchronization circuit 11. On the other hand, the frequency-multiplied pulse signal pls_fine, which is the output of the phase synchronization circuit 11, is frequency-divided into a 1 / N-times-frequency pulse signal by the 1 / N frequency divider 113 and then input to the phase comparator 110 in the same manner. The
The phase comparator 110 uses pls_cors as a reference wave, pls_fine is divided into 1 / N times frequency signals as a comparison wave, performs phase comparison between the reference wave and the comparison wave, and outputs the phase difference information as a pulse-like position. Expressed as a phase difference signal and output.
That is, if the logic of the reference wave and the comparison wave do not match, the logic of the phase difference signal is set to “H”, and if the logic of the reference wave and the comparison wave match, the logic is set to “L”. . Alternatively, when the logic of the reference wave and the comparison wave do not match, if the phase of the comparison wave is ahead of the reference wave, the logic is set to 'L', and the phase of the comparison wave is higher than the reference wave. The phase difference signal is generated by a method of setting the logic to “H” if it is delayed, and setting to the “high impedance state” when the logic of the reference wave and the comparison wave match.

This phase difference signal is input to the loop filter 111, which filters the AC component and outputs it.
The output of the loop filter 111 is input to the voltage controlled oscillator 112 as a control signal.
The voltage controlled oscillator 112 changes the oscillation frequency in accordance with the magnitude of the input control signal (voltage).
The output of the voltage controlled oscillator 112 is the output of the phase synchronization circuit 11, and the phase of the pulse signal obtained by dividing the output by the 1 / N frequency divider 113 to 1 / N times the frequency is the reference wave of the phase synchronization circuit 11. In order to perform feedback control so as to match the phase of the pulse signal pls_cors, the phase synchronization circuit 11 outputs a frequency multiplied pulse signal pls_fine that is obtained by multiplying pls_cors by N.
Subsequently, the frequency multiplication pulse signal pls_fine and the angle section identification signals sgn_cos and sgn_sin are input to the angle calculation unit 14 in the microprocessor 13.
The angle calculation unit 14 calculates the rotation angle θ using these three signals.

FIG. 8 is a block diagram showing a detailed configuration of the angle calculation unit 14 of the angle detection apparatus according to Embodiment 1 of the present invention.
The angle calculation unit 14 that is an angle calculation unit calculates the rotation angle of the resolver 2 based on the angle section display signal output by the section identification unit 12 and the final output of the phase synchronization circuit 11. The calculation unit 14 is a counting unit (a pulse counter 142 described later) that counts the number of pulses of the final output of the phase synchronization circuit 11 based on the angle interval display signal output by the interval identification unit 12 and the final output of the phase synchronization circuit 11. ) And angle information associated with the count value (the angle information storage memory 143 to be described later), and extraction means for extracting the angle information determined by the count value (controlled by the angle calculation unit 14). Means).

The specific operation of the angle calculation unit 14 is as follows.
In FIG. 8, the angle calculation unit 14 includes an angle calculation event trigger generator 141, a pulse counter 142, and an angle information storage memory 143.
The pulse counter 142 recognizes the rising edge of the frequency multiplied pulse signal pls_fine and counts up the counter.
Here, the counter is cleared in synchronization with the rising and falling edges of the angle section identification signals sgn_cos and sgn_sin.
That is, since the rising of the frequency multiplied pulse signal pls_fine obtained by multiplying the pulse signal pls_cors by N during the rising or falling timing of sgn_cos and sgn_sin is recognized and counted up, the count number is from zero to N−1. It is within the range of N stages up to.

The angle calculation event trigger generator 141 generates a trigger signal that triggers an angle calculation operation at an arbitrary time timing when the angle calculation unit 14 calculates the rotation angle θ.
Here, it is assumed that the angle calculation event trigger generator 141 generates a trigger signal in synchronization with the lapse of unit time of discrete time control performed by the microprocessor 13, but the angle calculation event trigger generator 141 The form is not limited to this, and the trigger signal may be generated at an arbitrary time timing.
When a trigger signal is generated by the angle calculation event trigger generator 141, the count value m of the pulse counter 142 at that time is taken out, and the count value m is transmitted to the angle information storage memory 143.

The angle information storage memory 143 divides the range from zero to 2π into four by the combination of the logic of the angle section identification signals sgn_cos and sgn_sin, and the count value of the pulse counter 142 for each section having the range of π / 2. And a table for mapping the rotation angle θ data on a one-to-one basis.
Considering the case where the angle section identification signal sgn_cos = “H” logic and sgn_sin = “L” logic, the range of the rotation angle θ is 3π / 2 to 2π.
Here, the data of the rotation angle θ corresponding to the case where the count value of the pulse counter 142 is zero is 3π / 2.
The rotation angle θ data corresponding to the count value of 1 is (3π / 2 + π / 2 × 1 / N) with π / 2 × 1 / N as an increment.
As the rotation angle θ increases, the count value also increases. The data of the rotation angle θ corresponding to the case where the count value is N−1 is (3π / 2 + π / 2 × (N−1) / N).
Further, at the timing when the rotation angle θ increases and the rising edge of the next frequency multiplication pulse signal pls_fine is recognized, the logic of the angle section identification signal sgn_sin is switched from “L” to “H” at the same time, and sgn_cos = “H” logic, sgn_sin = 'H' logic, and the count value of the pulse counter 142 is cleared to zero.

At this time, in the logic of the angle section identification signals sgn_sin and sgn_cos, the range of the rotation angle θ is 0 to π / 2, and the table referred to as the angle information storage memory 143 is switched to the target for this range.
That is, the data of the rotation angle θ corresponding to the case where the count value is zero is zero.
The rotation angle θ data corresponding to the count value of 1 is (0 + π / 2 × 1 / N) with π / 2 × 1 / N as an increment.
As the count value increases, the corresponding rotation angle θ data also increases, and the rotation angle θ data corresponding to the count value of N−1 is (0 + π / 2 × (N−1) / N). .

Similarly, when the angle section identification signals are sgn_cos = “L” logic and sgn_sin = “H” logic, the range of the rotation angle θ is π / 2 to π, and the table referred to as the angle information storage memory 143 Switches from π / 2 to π.
When the angle section identification signals are sgn_cos = “L” logic and sgn_sin = “L” logic, the range of the rotation angle θ is π to 3π / 2, and the table referred to as the angle information storage memory 143 is as follows: Switch from π to 3π / 2.

Now, when the count value m of the pulse counter 142 at that time is taken out by the trigger signal generated by the angle calculation event trigger generator 141 and transmitted to the angle information storage memory 143, the angle section identification signal sgn_cos at that time, Then, according to the logic combination of sgn_sin, an appropriate table corresponding to the range of the rotation angle θ is referred to, and the rotation angle θ is calculated by a one-to-one mapping.
Here, the rotation angle θ data when the count value is zero is stored at the start address of the memory area in which four tables in the π / 2 range are stored in the angle information storage memory 143, and the phase corresponding to each count value is stored. If the corner data is configured to be stored from the start address of the memory area to an address with the count value as an offset amount, the rotation angle θ based on the count value can be referred to at high speed.

That is, the direct memory access (DMA) function provided in the microprocessor 13 is utilized, and the count value of the pulse counter 142 is transmitted to the angle information storage memory 143 with the trigger signal generation of the angle calculation event trigger generator 141 as a starting point. A series of operations from the start address of the memory to the calculation of the phase angle data stored at the address with the count value as an offset amount is performed in advance without depending on the microprocessor software operation (microcode execution). It can be realized as a hardware operation of the set microprocessor.
This is because the count value of the pulse counter 142 is set in advance so as to be the offset amount of the memory reference address in the direct memory access (DMA) function, and the angle calculation event trigger generator 141 generates the trigger signal directly. This is realized by operating the memory access function and extracting the memory contents at the address corresponding to the count value of the trigger signal generation timing.

The angle detection apparatus of the present invention operates according to the above procedure.
Here, the angle information storage memory 143 stores the rotation angle θ data in a one-to-one correspondence from the count range zero to N−1 of the pulse counter 142 corresponding to the angle range π / 2. Thus, 4 × N pieces of angle data are obtained in the range 0 to 2π of the rotation angle θ.
For this reason, the resolution of the rotation angle θ is expressed as 2π / (4 × N) = π / (2 × N). Therefore, the resolution of the rotation angle θ is improved by increasing the multiplication factor N of the frequency multiplication pulse signal pls_fine.
For example, when expressing 2π from the range of the rotation angle θ in the range of 1,000, the multiplication factor N may be set to 250.

According to the present invention, since a division operation is not required for calculating the rotation angle θ, even when combined with a discrete time control device while reducing the processing time, the control response and accuracy can be reduced at a low cost. Maintenance or improvement can be achieved.
In addition, by setting the multiplication factor of the frequency-multiplied pulse signal based on the required resolution of the rotation angle θ, redundant data conversion processing becomes unnecessary in the calculation process of the rotation angle θ based on the number of measured pulses, and the processing time is further increased. It is possible to provide an angle detection device in which is shortened.

In the first embodiment, the case where the present invention is applied to a resolver having a single-phase excitation and two-phase output has been illustrated. However, the application target is not necessarily limited to this type of resolver.
For example, when applied to a phase modulation resolver having two-phase excitation and one-phase output, one pulse signal pls_cors is generated and output based on an output signal for one phase, and the pulse of the frequency-multiplied pulse signal pls_fine obtained by multiplying this by N. In addition to counting up the number, the rotation angle θ may be calculated by referring to the angle information storage memory using a count value at the zero cross timing of the two-phase excitation signal.

In the first embodiment, the table stored and referred to in the angle information storage memory 143 is divided into four ranges by combining the range 0 to 2π of the rotation angle θ with the combination of the angle section identification signals sgn_cos and sgn_sin logic. Although an example is shown in which / 2 is configured as one block, the configuration of the table to be referred to is not necessarily limited to this form.
For example, the angle interval identification signal sgn_cos is converted into a pulse signal pls_cors, and a frequency multiplied pulse signal pls_fine is generated by multiplying it by N. The pulse counter 142 counts the number of pulses and recognizes the rising edge of the angle interval identification signal sgn_cos. Then, the count value may be cleared.
In such a configuration, the table referred to as the angle information storage memory 143 is one memory having N elements corresponding to the range of the rotation angle θ from zero to 2π without being divided according to the angle section. It can be realized as a block.

  In the angle detection device according to the first embodiment of the present invention, the frequency of a signal obtained by demodulating the modulation signal of the resolver is multiplied to obtain a frequency-multiplied pulse signal, and the pulse count number of this pulse signal and the demodulated signal The modification which can detect rotation angle (theta) based on is also included.

Embodiment 2. FIG.
FIG. 9 is a block diagram showing an overall configuration of a rotational coordinate calculation device using the angle detection device according to the second embodiment of the present invention.
The rotating coordinate calculation device shown in FIG. 9 includes a coordinate calculation unit 15 instead of the angle calculation unit 14 in the microprocessor 13 of the angle detection device according to the first embodiment shown in FIG.
Since the other configuration is the same as that of the angle detection device shown in FIG. 1, the same or conceivable components are denoted by the same reference numerals, and the description thereof is omitted.
Hereinafter, the rotational coordinate arithmetic apparatus according to the second embodiment of the present invention will be described in different points from the angle detection apparatus according to the first embodiment.

As shown in FIG. 9, based on the cosine function modulation signal mod_cos of the resolver 2, the cosine function demodulation signal dem_cos demodulated by removing the carrier component from the sine function modulation signal mod_sin, and the sine function demodulation signal dem_sin, Angular section identification signals sgn_cos and sgn_sin are generated from the code.
Next, after the binary logic of sgn_cos and sgn_sin is synthesized by exclusive OR to obtain one pulse signal pls_cors, the frequency-multiplied pulse signal pls_fine whose frequency is multiplied by N by the phase synchronization circuit 11 is obtained. Generated.

Subsequently, the frequency multiplication pulse signal pls_fine, the angle section identification signal sgn_cos, and sgn_sin are input to the coordinate calculation unit 15 in the microprocessor 13.
The coordinate calculation unit 15 calculates rotational coordinates using these pieces of information.
FIG. 10 is a block diagram showing a detailed configuration of a coordinate calculation unit 15a that performs a rotation coordinate calculation around the origin on a two-dimensional plane as a rotation coordinate calculation in the rotation coordinate calculation apparatus according to Embodiment 2 of the present invention.

The coordinate calculation unit 15 a is a rotation coordinate calculation unit that calculates a rotation coordinate based on the rotation angle of the resolver 2 based on the angle section display signal output by the section identification unit 12 and the final output of the phase synchronization circuit 11.
Further, the coordinate calculation unit 15a is a counting unit (described later) that counts the number of pulses of the final output of the phase synchronization circuit 11 based on the angle interval display signal output by the interval identification unit 12 and the final output of the phase synchronization circuit 11. And a memory means (sine information storage memory 153a and cosine information storage memory 153b described later) for storing sine information and cosine information associated with the count value, and sine information and cosine information determined by the count value. And coordinate calculation means (coordinate rotation means 155a described later) for calculating coordinates after rotation using the coordinates before rotation.

As shown in FIG. 10, the coordinate calculation unit 15a includes a coordinate calculation event trigger generator 151, a pulse counter 152, a sine information storage memory 153a, a cosine information storage memory 154, and a coordinate rotation unit 155a.
Similar to the movement of the pulse counter 142 in the first embodiment, the pulse counter 152 recognizes the rising edge of the frequency-multiplied pulse signal pls_fine and counts up.
In addition, the pulse counter 152 is cleared in synchronization with the rising or falling edges of the angle section identification signals sgn_cos and sgn_sin.

The coordinate calculation event trigger generator 151 generates a trigger signal that triggers the arbitrary timing when the coordinate calculation unit 15 performs the rotation coordinate calculation.
When a trigger signal is generated by the coordinate calculation event trigger generator 151, the count value m of the pulse counter 152 at that time is taken out and transmitted to the sine information storage memory 153a and the cosine information storage memory 154.

The sine information storage memory 153a divides the range from zero to 2π into four by the combination of the angle interval identification signals sgn_cos and sgn_sin logic, and the count value of the pulse counter 152 for each interval having π / 2 as the range, The table is configured as a table that maps the corresponding sine value data of the rotation angle θ on a one-to-one basis.
Similarly, the cosine information storage memory 154 is configured as a table for one-to-one mapping of the count value of the pulse counter 152 and the corresponding cosine value data of the rotation angle θ for each section having a range of π / 2. It has been done.

Considering the case where the angle section identification signal sgn_cos = “H” logic and sgn_sin = “L” logic, the range of the rotation angle θ is 3π / 2 to 2π.
Here, the rotation angle θ corresponding to the case where the count value of the pulse counter 152 is zero is 3π / 2, and the sine value thereof is sin (3π / 2) = − 1.0. The sine information data of 153a is -1.0.
The rotation angle θ corresponding to the case where the count value is 1 is (3π / 2 + π / 2 × 1 / N) with π / 2 × 1 / N as an increment, and the sine information data in the sine information storage memory 153a is The sine value sin (3π / 2 + π / 2 × 1 / N) is stored.
As the rotation angle θ increases, the count value also increases, and the rotation angle θ corresponding to the case where the count value is N−1 is (3π / 2 + π / 2 × (N−1) / N). As the sine information data in the sine information storage memory 153a, the sine value sin (3π / 2 + π / 2 × (N−1) / N) is stored.

Further, since the cosine value of the rotation angle θ corresponding to the case where the count value of the pulse counter 152 is zero is cos (3π / 2) = 0.0, the cosine information data in the cosine information storage memory 154 is 0.0. It is.
When the count value is 1, the corresponding rotation angle θ is (3π / 2 + π / 2 × 1 / N), and the cosine information data in the cosine information storage memory 154 includes its cosine value cos (3π / 2 + π / 2). × 1 / N) is stored.
Further, the rotation angle θ corresponding to the case where the count value increases to N−1 is (3π / 2 + π / 2 × (N−1) / N), and is used as cosine information data in the cosine information storage memory 154. The cosine value cos (3π / 2 + π / 2 × (N−1) / N) is stored.

  Further, at the timing when the rotation angle θ increases and the rising edge of the next frequency multiplication pulse signal pls_fine is recognized, the logic of the angle section identification signal sgn_sin is switched from “L” to “H” at the same time, and sgn_cos = “H” logic. And sgn_sin = 'H' logic, and the count value of the pulse counter 152 is cleared to zero.

At this time, in the logic of the angle section identification signals sgn_sin and sgn_cos, the range of the rotation angle θ is 0 to π / 2, and the table referred to as the sine information storage memory 153a and the cosine information storage memory 154 covers this range. Switch to things.
That is, since the rotation angle θ corresponding to the count value being zero is zero, the sine value 0.0 is stored as the sine information data in the sine information storage memory 153a, and the cosine information in the cosine information storage memory 154 is stored. As data, the cosine value 1.0 is stored.

The rotation angle θ corresponding to the case where the count value is 1 is (0 + π / 2 × 1 / N) with an increment of π / 2 × 1 / N, and the sine information data in the sine information storage memory 153a is its sine value. sin (0 + π / 2 × 1 / N) is stored, and the cosine value cos (0 + π / 2 × 1 / N) is stored as the cosine information data in the cosine information storage memory 154.
As the rotation angle θ increases, the corresponding count value also increases, and the rotation angle θ corresponding to the case where the count value is N−1 is (0 + π / 2 × (N−1) / N).
As the sine information data in the sine information storage memory 153a, the sine value sin (0 + π / 2 × (N−1) / N) is stored. As the cosine information data in the cosine information storage memory 154, the cosine value cos ( 0 + π / 2 × (N−1) / N) is stored.

In the same manner as above, when the angle section identification signals are sgn_cos = “L” logic and sgn_sin = “H” logic, the range of the rotation angle θ is π / 2 to π, and is referred to as the sine information storage memory 153a. The table to be switched from π / 2 to one storing sine values targeting π, and the table referred to as the cosine information storage memory 154 stores cosine values targeting π / 2 to π. Switch to
When the angle section identification signals are sgn_cos = “L” logic and sgn_sin = “L” logic, the range of the rotation angle θ is π to 3π / 2, and the table referred to as the sine information storage memory 153a is The table is switched from π to 3π / 2, and the table referred to as the cosine information storage memory 154 is switched from π to 3π / 2.

  Now, the count value m of the pulse counter 152 at that time is taken out by the trigger signal generated by the coordinate calculation event trigger generator 151 and transmitted to the sine information storage memory 153a and the cosine information storage memory 154, respectively. In accordance with the combination of the logic of the angle interval identification signals sgn_cos and sgn_sin, an appropriate table corresponding to each π / 2 range of the rotation angle θ is referred to, and a sine value sin (θ), cosine is obtained by each one-to-one mapping. The value cos (θ) is calculated.

  Here, in the sine information storage memory 153a and the cosine information storage memory 154, the sine information corresponding to the rotation angle θ when the count value is zero is stored at the top address of each of the memory areas storing the four tables in the π / 2 range. If the sine information and cosine information of the phase angle θ corresponding to each count value is stored in an address with the count value as an offset amount from the start address of each memory block, the count value is Based on the direct memory access (DMA) function provided in the microprocessor 13, the sine information and cosine information of the rotation angle θ can be referred to at high speed.

  In the same manner as the operation procedure of the direct memory access function described in the first embodiment, the count value of the pulse counter 152 is set in advance so as to be the offset amount of the memory reference address in the direct memory access function. This is realized by operating the direct memory access function at the trigger signal generation timing of the event trigger generator 151 and taking out the memory contents at the address corresponding to the count value of the trigger signal generation timing.

Next, together with the sine value output from the sine information storage memory 153a and the cosine value output from the cosine information storage memory 154, the position data ux, uy on the two-dimensional plane, which is the rotational coordinate calculation target, are sent to the coordinate rotation means 155a. Entered.
The coordinate rotation unit 155a performs a rotation coordinate calculation around the origin on the two-dimensional plane shown in Expression 1 based on these input data ux and ui, and outputs post-rotation position data jx and jy.

With the above procedure, the rotational coordinate calculation device of the present invention operates.
Here, the sine information storage memory 153a and the cosine information storage memory 154 have a one-to-one correspondence with the sine value of the rotation angle θ with respect to the count range zero to N−1 of the pulse counter 152 corresponding to the angle range π / 2. Since the cosine value is stored, 4 × N pieces of sine value information data and cosine value information data are provided in the range 0 to 2π of the rotation angle θ.
Therefore, the resolution of the sine value information data and the cosine value information data is represented as 2π / (4 × N) = π / (2 × N).
Therefore, the resolution of the sine value information data and the cosine value information data can be improved by increasing the multiplication factor N of the frequency multiplication pulse signal pls_fine.

As described above, according to the rotational coordinate calculation apparatus according to the second embodiment of the present invention, in calculating the trigonometric function value used for the calculation in the rotation coordinate calculation, the process of calculating the rotation angle represented by the data is not performed. It is possible to provide a rotating coordinate calculation device that shortens the time required for calculation calculation processing.
In addition, by setting the multiplication factor of the frequency-multiplied pulse signal based on the required resolution of the trigonometric function value used for the rotation coordinate calculation, redundant data conversion processing becomes unnecessary in the calculation process of the trigonometric function value, and the processing time is further reduced. It can be shortened.

In the rotational coordinate arithmetic unit according to Embodiment 2 of the present invention, the frequency of a signal obtained by demodulating the resolver modulation signal is multiplied to obtain a frequency-multiplied pulse signal, and the pulse count number and demodulation of this pulse signal are obtained. A modification example is also included in which the rotation information data used for the rotation coordinate calculation can be calculated based on the signal without calculating the rotation angle θ represented by the data.
The calculation of the rotation coordinate is not limited to the rotation around the origin on the two-dimensional plane, but can also be applied to the coordinate rotation in the integer L-dimensional space.

Embodiment 3 FIG.
The configuration of the rotation coordinate calculation device according to the third embodiment of the present invention is basically the same as the configuration of the rotation coordinate calculation device according to the second embodiment (see FIG. 9). Is different.
Hereinafter, the configuration of the coordinate calculation unit of the rotary coordinate calculation device according to Embodiment 3 of the present invention and the contents of the calculation process will be described.

Similarly to the second embodiment, the frequency multiplication pulse signal pls_fine, the angle section identification signal sgn_cos, and sgn_sin are input to the coordinate calculation unit 15b (indicated by reference numeral 15a in FIG. 9) in the microprocessor 13.
The coordinate calculation unit 15b uses these pieces of information to calculate the coordinate conversion from the three-phase AC coordinate to the rotation orthogonal coordinate (d, q), or the coordinate conversion from the rotation orthogonal coordinate (d, q) to the three-phase AC coordinate. I do.

FIG. 11 is a diagram showing the configuration of the coordinate calculation unit of the rotational coordinate calculation apparatus according to Embodiment 3 of the present invention.
The coordinate calculation unit 15 b is a rotation orthogonal coordinate calculation unit that calculates a rotation orthogonal coordinate based on the rotation angle of the resolver 2 based on the angle interval display signal output by the interval identification unit 12 and the final output of the phase synchronization circuit 11. is there.
The coordinate calculation unit 15b is a counting unit (described later) that counts the number of pulses of the final output of the phase synchronization circuit 11 based on the angle interval display signal output by the interval identification unit 12 and the final output of the phase synchronization circuit 11. Rotating using a pulse counter 152), memory means for storing sine information associated with the count value (sine information storage memory 153b described later), sine information determined by the count value, and three-phase coordinates before rotation Coordinate calculating means for calculating orthogonal coordinates (dq coordinate calculating means 155b described later).

The operation of the coordinate calculation unit 15b is as follows.
As shown in FIG. 11, the coordinate calculation unit 15b performs coordinate conversion calculation from three-phase alternating current coordinates to rotation orthogonal coordinates as coordinate conversion calculation, coordinate calculation event trigger generator 151, pulse counter 152, sine information storage. A memory 153b and dq coordinate calculation means 155b are provided.
The coordinate calculation event trigger generator 151 and the pulse counter 152 are the same as those provided in the coordinate calculation unit 15a of the rotary coordinate calculation device according to the second embodiment.

The pulse counter 152 counts up by recognizing the rising edge of the frequency-multiplied pulse signal pls_fine, and the counter is cleared in synchronization with each rising edge or falling edge of the angle section identification signals sgn_cos and sgn_sin. .
When a trigger signal is generated by the coordinate calculation event trigger generator 151, the count value m of the pulse counter 152 at that time is taken out and transmitted to the sine information storage memory 153b.

  The sine information storage memory 153b divides the range from 0 to 2π into four by the combination of the angle interval identification signals sgn_cos and sgn_sin logic, and the count value of the pulse counter 152 and the count value of each interval in the range of π / 2. Four sine information (√2) sin θ, (√2) sin (θ + π / 3), (√2) sin (θ + π / 2), (√2) sin (θ + 5π / 6) relating to the rotation angle θ corresponding to Each table is configured as a one-to-one mapping table.

Considering the case where the angle section identification signal sgn_cos = “H” logic and sgn_sin = “L” logic, the range of the rotation angle θ is 3π / 2 to 2π.
Here, the rotation angle θ corresponding to the case where the count value of the pulse counter 152 is zero is 3π / 2, and the data for the sine information (√2) sin θ is (√2) sin (3π / 2) = −, respectively. The data for (√2), sine information (√2) sin (θ + π / 3) is (√2) sin (3π / 2 + π / 3) = − (√2) / 2, sine information (√2) sin (θ + π / 2) data is (√2) sin (3π / 2 + π / 2) = 0, sine information (√2) sin (θ + 5π / 6) data is (√2) sin (3π / 2 + 5π / 6) = √ (3/2).

  The rotation angle θ corresponding to the case where the count value is 1 is (3π / 2 + π / 2 × 1 / N) with an increment of π / 2 × 1 / N, and the data for sine information (√2) sin θ is (√2) sin (3π / 2 + π / 2 × 1 / N), sine information (√2) data for sin (θ + π / 3) is (√2) sin (3π / 2 + π / 2 × 1 / N + π / 3) The data for sine information (√2) sin (θ + π / 2) is (√2) sin (3π / 2 + π / 2 × 1 / N + π / 2), and the data for sine information (√2) sin (θ + 5π / 6) is (√2) sin (3π / 2 + π / 2 × 1 / N + 5π / 6).

  As the rotation angle θ increases, the count value also increases, and the rotation angle θ corresponding to the case where the count value is N−1 is (3π / 2 + π / 2 × (N−1) / N). The data for sine information (√2) sin θ is (√2) sin (3π / 2 + π / 2 × (N−1) / N), and the data for sine information (√2) sin (θ + π / 3) is (√ 2) The data for sin (3π / 2 + π / 2 × (N−1) / N + π / 3) and sine information (√2) sin (θ + π / 2) is (√2) sin (3π / 2 + π / 2 × (N −1) / N + π / 2) and data for sine information (√2) sin (θ + 5π / 6) are (√2) sin (3π / 2 + π / 2 × (N−1) / N + 5π / 6).

  Further, when the rotation angle θ is increased and the rising edge of the next frequency multiplied pulse signal pls_fine is recognized, the logic of the angle section identification signal sgn_sin is switched from 'L' to 'H' at the same time, and sgn_cos = 'H' logic, sgn_sin = 'H' logic, and the count value of the pulse counter 152 is cleared to zero.

At this time, in the logic of the angle section identification signals sgn_sin and sgn_cos, the range of the rotation angle θ is 0 to π / 2, and the table referred to as the sine information storage memory 153b is switched to the target of this range.
Similarly, when the angle section identification signals are sgn_cos = “L” logic and sgn_sin = “H” logic, the range of the rotation angle θ is π / 2 to π, and is referred to as the sine information storage memory 153b. The table to be switched switches from π / 2 to one storing sine information for π.
When the angle section identification signals are sgn_cos = “L” logic and sgn_sin = “L” logic, the range of the rotation angle θ is π to 3π / 2, and the table referred to as the sine information storage memory 153b is as follows: Switching from π to 3π / 2 is stored.

  Now, when the count value m of the pulse counter 152 at that time is taken out by the trigger signal generated by the coordinate calculation event trigger generator 151 and transmitted to the sine information storage memory 153b, the angle interval identification signal sgn_cos, According to the logic combination of sgn_sin, an appropriate table corresponding to the range of the rotation angle θ is referred to, and by each one-to-one mapping, a sine value (√2) sin θ, (√2) sin (θ + π / 3) , (√2) sin (θ + π / 2), (√2) sin (θ + 5π / 6).

  Here, sine value information corresponding to the rotation angle θ when the count value is zero is stored at the top address of each memory area in which four tables in the π / 2 range are stored in the sine information storage memory 153b. When the sine information of the phase angle θ corresponding to the count value is stored at an address with the count value as an offset amount from the start address of each memory block, the rotation angle based on the count value is the same as in the second embodiment. The sine value and cosine value of θ can be referred to at high speed by utilizing the direct memory access (DMA) function provided in the microprocessor 13.

Next, each sine information (√2) sin θ, (√2) sin (θ + π / 3), (√2) sin (θ + π / 2), (√2) sin (θ + 5π) output from the sine information storage memory 153b. / 6), the three-phase information ju, jv, which is the coordinate transformation calculation target, is input to the dq coordinate calculation means 155b.
Based on these input data, the dq coordinate calculation means 155b performs the coordinate conversion calculation from the three-phase coordinates shown in Equation 2 to the rotation orthogonal coordinates (d, q), and outputs the data jd and jq after the coordinate conversion. .

As described above, according to the rotational coordinate calculation device according to Embodiment 3 of the present invention, the trigonometric function value used for the calculation in the coordinate conversion calculation does not go through the process of calculating the rotation angle represented by the data. It is possible to provide a rotational coordinate calculation device that shortens the time required for processing.
In the rotational coordinate arithmetic unit according to the third embodiment of the present invention, the frequency of a signal obtained by demodulating the modulation signal of the resolver is multiplied to obtain a frequency-multiplied pulse signal, and the pulse count number and demodulation of this pulse signal are obtained. There is also included a modified example in which a trigonometric function value used for coordinate transformation calculation can be calculated based on a signal without calculating a rotation angle represented by data.

  Further, the coordinate conversion may be any of conversion from three-phase coordinates to rotation orthogonal coordinates (d, q) and conversion from rotation orthogonal coordinates (d, q) to three-phase coordinates. Is converted from three-phase AC information based on three inputs, or based on the input of any two phases of the three phases. In any case, the time required for processing can be shortened.

It is a block diagram which shows the whole structure of the angle detection apparatus which concerns on Embodiment 1 of this invention. A waveform example of the rotation angle θ, the excitation signal ref_sig, the cosine function modulation signal mod_cos, and the sine function modulation signal mod_sin with respect to the common horizontal axis is shown. Waveform examples of the cosine function-like modulation signal mod_cos, the cosine function-like demodulation signal dem_cos, the sine function-like modulation signal mod_sin, and the sine function-like demodulation signal dem_sin when the rotation angle θ changes as shown in FIG. Indicates. The division characteristics of the cosine function demodulated signal dem_cos and the sine function demodulated signal dem_sin are shown. The relationship between the output logic of the angle section identification signal, the sign of the cosine function demodulated signal dem_cos, and the sign of the sine function demodulated signal dem_sin is shown. A waveform example of the rotation angle θ with respect to the time axis which is a common horizontal axis, angle section identification signals sgn_cos and sgn_sin, and a pulse signal pls_cors is shown. It is a block diagram which shows the detailed structure of the phase synchronizing circuit 11 of the angle detection apparatus which concerns on Embodiment 1 of this invention. It is a block diagram which shows the detailed structure of the angle calculation part 14 of the angle detection apparatus which concerns on Embodiment 1 of this invention. It is a block diagram which shows the whole structure of the rotation coordinate calculating apparatus using the angle detection apparatus which concerns on Embodiment 2 of this invention. It is a block diagram which shows the detailed structure of the coordinate calculating part 15a which performs the rotation coordinate calculation around the origin on a two-dimensional plane as rotation coordinate calculation in the rotation coordinate calculation apparatus which concerns on Embodiment 2 of this invention. It is a figure which shows the structure of the coordinate calculating part of the rotation coordinate calculating apparatus which concerns on Embodiment 3 of this invention.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 1 Angle detection apparatus, 2 Resolver, 3 Excitation circuit, 4 Input interface circuit, 9 Demodulation means, 10 Pulse signal generator, 11 Phase synchronization circuit, 12 Section identification means, 13 Microprocessor, 14 Angle calculation part, 15a Two-dimensional plane A coordinate calculation unit for calculating rotation coordinates around the origin, 15b a coordinate calculation unit for calculating rotation coordinates from three-phase coordinates to rotation orthogonal coordinates, 21 primary winding (excitation), 22 secondary winding (COS), 23 secondary winding (SIN), 24 rotor, 110 phase comparator, 111 loop filter, 112 voltage controlled oscillator, 113 1 / N frequency divider, 141 angle calculation event trigger, 142 pulse counter, 143 angle information storage memory, 151 Coordinate operation event trigger, 152 pulse counter, 153a, 153b, sine information storage memory 154 cosine information storage memory, 155a coordinate rotation means, 155b dq coordinate calculation means.

Claims (6)

An angle detection device that detects a rotation angle of a resolver that receives an excitation signal of a periodic function and outputs an amplitude modulation signal or a phase modulation signal using the excitation signal as a carrier wave,
Demodulation means for demodulating the resolver modulation signal;
Identifying an angle section based on the demodulated signal demodulated by the demodulating means, and outputting an angle section display signal representing the angle section according to the identification result;
A pulse signal generating means for generating a pulse signal based on the identification result of the section identifying means;
A phase synchronization means for obtaining a pulsed final output by comparing the phase of the reference signal and a comparison signal obtained by multiplying the final output, using the pulse signal generated from the pulse generating means as a reference signal;
An angle calculation means for calculating the rotation angle of the resolver based on the angle section display signal output by the section identification means and the final output of the phase synchronization means,
The angle calculation means includes
Counting means for counting the number of pulses of the final output of the phase synchronization means based on the angle interval display signal output by the section identification means and the final output of the phase synchronization means;
Memory means for storing angle information associated with the count value;
An angle detection apparatus comprising: extraction means for extracting angle information determined by the count value.   2. The angle detection device according to claim 1, wherein the multiplication factor in the phase synchronization means is determined based on a resolution of data of the rotation angle. A rotation coordinate calculation device that calculates a rotation coordinate based on a rotation angle of a resolver that receives an excitation signal of a periodic function and outputs an amplitude modulation signal or a phase modulation signal using the excitation signal as a carrier,
Demodulation means for demodulating the resolver modulation signal;
Identifying an angle section based on the demodulated signal demodulated by the demodulating means, and outputting an angle section display signal representing the angle section according to the identification result;
A pulse signal generating means for generating a pulse signal based on the identification result of the section identifying means;
A phase synchronization means for obtaining a pulsed final output by comparing the phase of the reference signal and a comparison signal obtained by multiplying the final output, using the pulse signal generated from the pulse generating means as a reference signal;
A rotation coordinate calculation means for calculating a rotation coordinate based on the rotation angle of the resolver based on the angle section display signal output by the section identification means and the final output of the phase synchronization means;
The rotational coordinate calculation means includes
Counting means for counting the number of pulses of the final output of the phase synchronization means based on the angle interval display signal output by the section identification means and the final output of the phase synchronization means;
Memory means for storing sine information and cosine information associated with the count value;
A rotation coordinate calculation device comprising: coordinate calculation means for calculating coordinates after rotation using sine information and cosine information determined by the count value and coordinates before rotation.   4. The rotation coordinate calculation apparatus according to claim 3, wherein the calculation of the rotation coordinate by the rotation coordinate calculation means is an operation for rotating coordinates in an integer L-dimensional space including a two-dimensional plane. A rotation coordinate calculation device that calculates a rotation coordinate based on a rotation angle of a resolver that receives an excitation signal of a periodic function and outputs an amplitude modulation signal or a phase modulation signal using the excitation signal as a carrier,
Demodulation means for demodulating the resolver modulation signal;
Identifying an angle section based on the demodulated signal demodulated by the demodulating means, and outputting an angle section display signal representing the angle section according to the identification result;
A pulse signal generating means for generating a pulse signal based on the identification result of the section identifying means;
A phase synchronization means for obtaining a pulsed final output by comparing the phase of the reference signal and a comparison signal obtained by multiplying the final output, using the pulse signal generated from the pulse generating means as a reference signal;
A rotation orthogonal coordinate calculation means for calculating a rotation orthogonal coordinate based on the rotation angle of the resolver based on the angle interval display signal output by the interval identification means and the final output of the phase synchronization means;
The rotation orthogonal coordinate calculation means includes:
Counting means for counting the number of pulses of the final output of the phase synchronization means based on the angle interval display signal output by the section identification means and the final output of the phase synchronization means;
Memory means for storing sine information associated with the count value;
A rotation coordinate calculation device comprising: coordinate calculation means for calculating rotation orthogonal coordinates using sine information determined by the count value and three-phase coordinates before rotation.   The rotational coordinate calculation according to any one of claims 3 to 5, wherein the multiplication factor in the phase synchronization means is determined based on a resolution of rotation information data including a trigonometric function value at the time of the rotation coordinate calculation. apparatus.

Images