JP3508718B2 - Rotation angle detector - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION [0001] BACKGROUND OF THE INVENTION,Dynamic motor
Rotation angle detector that measures the rotation angle of the rotor with respect to the stator
About the installation. [0002] 2. Description of the Related Art Conventionally, for example, Japanese Patent Application Laid-Open No. 9-28115
No. 8, as shown in the input sinusoidal signal
At the specified rate and sample
A / D converts the sampled value of the input sine wave signal
From the plurality of A / D converted sample values, zero
The frequency of the input sinusoidal signal derived based on the cross detection
Multiple sample values near the peak value based on the period
And averaged multiple sample values
Specified by upper and lower limits determined based on the value
Range and exclude sample values that fall outside the range.
Calculating the amplitude value of the input sinusoidal signal is not done.
Was. [0003] In addition, in resolvers such as electric motors,
And a first winding is provided on one of the stator and the rotor.
And the other is orthogonal to the first winding facing the first winding.
Second and third windings, and the first winding has an analog sine wave
Applying a signal to each other induces π in the second and third windings.
Sine wave signal shifted in phase by / 2, ie, SIN phase signal
Of the rotor with respect to the stator based on the
Detecting the rotation angle θ is well known. This place
For example, phase sensing detector (synchronous detection circuit), integration
, A voltage controlled oscillator, etc., the second and third
Input signal from winding to form signal synchronized with rotation angle θ
A closed loop circuit (a kind of phase locked loop)
Circuit) to detect the rotation angle θ.
You. [0004] SUMMARY OF THE INVENTION
Of the input sine wave signal without being affected by noise.
It is possible to detect the amplitude value, but the calculated
The error due to the sampling rate and A / D conversion
The difference is included and their sampling rate and A
Of the amplitude value of the input sine wave signal
The detection accuracy cannot be improved. In the latter prior art, the rotation angle
The problem that the circuit size for detecting θ becomes large
And a high-speed clock signal is needed
There is also a problem. [0006] SUMMARY OF THE INVENTION The present invention addresses the problems described above.
The premise is that it is measured
SignalIs positiveChord-likeInTherefore, the frequency of the sine wave signal is already
It applies when you are knowledgeable. And the present invention
The purpose of this is to use this assumption to
Of the sine wave component having the predetermined frequency contained inSimple widthsingle
And highly accurate detectionThen, using the detected amplitude,Movement
Rotation angle of rotor with respect to stator such as motorSimplifiedSimple and
Measuring with high accuracyInYou. [0007] [0008] [0009] [0010] [0011] [0012] [0013] [0014] [0015] [0016] [0017] [0018] [0019] [0020] [0021] [0022] [0023] [0024] [0025]To achieve the above objectives,Configuration of the present invention
The feature above is that the first volume is applied to either the stator or the rotor.
Wire, and the first winding on the other of the stator and the rotor.
And the second and third windings which are orthogonal to each other
Rotation to detect the rotation angle of the rotor with respect to the stator
In the angle detection device, a predetermined reference analog sine wave signal is generated.
A reference signal applying means for applying to the first winding;
The voltage induced by the winding as the first measured analog signal
Input and included in the input first measured analog signal
A sine wave component having the same frequency as the reference analog sine wave signal
First amplitude deriving means for deriving the amplitude of
Input voltage signal as a second measurement analog signal
The base signal included in the input second measurement analog signal.
Amplitude of sine wave component of same frequency as quasi-analog sine wave signal
And second amplitude deriving means for deriving the first and second amplitudes
The fixed based on each amplitude calculated by the deriving means
Rotation angle calculating means for calculating the rotation angle of the rotor with respect to the rotor;
That you have. In this case, the reference analog sine wave signal and
If the turning angles are Ao · sin2πft and θ, respectively,
And the voltage induced by the third winding, ie, the first and second measurements
The constant analog signals are Ax, Ao, sinθ, sin2πf, respectively.
t and Ax · Ao · cosθ · sin2πft. No.1st and 1st
The two amplitude deriving means is configured to output the first and second measurement analogs.
Signal Ax · Ao · sinθ · sin2πft, Ax · Ao · cosθ · sin2
πft, respectively, and the reference analog sine wave signal A
The amplitude Ax of the sine wave component having the same frequency f as o · sin2πft
Ao · sinθ and Ax · Ao · cosθ are derived. And the rotation angle
The calculating means calculates the amplitudes Ax, Ao, sinθ, Ax, Ao, cos
Calculate the rotation angle of the rotor with respect to the stator based on θ
I do. For example, each amplitude Ax · Ao · sinθ, Ax · Ao · cosθ ratio
Arc tangent of the value sinθ / cosθ (tan^-1)
By calculating, the rotation angle θ is calculated. As a result, according to the present invention, the stator
The rotation angle θ of the rotating rotor can be detected with a simple configuration.
Become so. According to the present invention, each of the amplitudes A
To detect x ・ Ao ・ sinθ and Ax ・ Ao ・ cosθ,
First and second amplitude deriving meansNextAsConfusedTo achieve. That is, the first amplitude deriving means includes:
Samples the input first measured analog signal at a predetermined rate
1st measurement analog sampled and sampled
First A / D conversion means for sequentially A / D converting signals;
The predetermined number of A / D converted by the first A / D converting means
Using sampling data and the reference analog
A sine wave function or cosine wave function of the same frequency as the sine wave signal
Applying the least squares method as the basis function,
By calculating the number, the first measured analog signal
Included and at the same frequency as the reference analog sine wave signal
And first amplitude calculating means for calculating the amplitude of the sine wave component of
And inputting the second amplitude deriving means to the
The second measured analog signal at a predetermined rate.
And the second sampled analog signal
Second A / D conversion means for sequentially performing A / D conversion;
Of the predetermined number of A / D converted by the A / D conversion means.
Using the reference analog positive
Based on a sine or cosine function at the same frequency as the sinusoidal signal
Apply the least-squares method as the base function to obtain the coefficients of the same basis function
Is calculated to include in the second measured analog signal
Of the same frequency as the reference analog sine wave signal
A second amplitude calculator for calculating the amplitude of the sine wave component
it can. According to this,The mostBy adopting the small square method,
Influence of sampling rate and error due to A / D conversion
Each of the amplitudes Ax · Ao · sinθ and Ax · Ao · cosθ
Is detected with high accuracy, so the detection accuracy of the rotation angle θ
improves. Further, aboveNotationThe basic features and basis functions
do it, BasisOf the same frequency as the frequency of the quasi-analog sine wave signal.
Sine wave function and / or cosine wave function are adopted
Is the basis function corresponding to the DC component or 2
Sine wave function with the same frequency as the harmonic frequency
Using the number and / or cosine wave function,
Same as the coefficient of the corresponding basis function and / or the frequency of the same harmonic
Each consisting of a sine wave function and / or a cosine wave function of one frequency
It is preferable to calculate each coefficient of the basis function. In this case, the basis function corresponding to the DC component
Coefficient is the DC offset component of the input analog signal.
Sine wave function of the same frequency
Each coefficient of each basis function consisting of a number and / or a cosine wave function is
Corresponds to harmonic noise. Therefore, for each of these coefficients
On the basis of, TimesAbnormalities in measurement such as turning angles can also be detected
Become like [0033] DETAILED DESCRIPTION OF THE INVENTION a. Basic principle According to the present invention, as shown in FIG.
Although it is a known sine waveform (cosine waveform), DC noise, harmonics
Wave noise, phase lag or lead with respect to reference sine wave signal
Input a sinusoidal detection signal including
Using the least squares method employing sine and / or cosine waves
The input detection signal is analyzed by the
Measures various physical quantities such as temperature, force, displacement, etc. expressed as
To do. Before describing specific embodiments
The sine wave function and cosine wave function used in the present invention are
The least square method as a number will be described. In the least square method, an arbitrary waveform signal y (t) is
A basis function φ that is linearly independent_jUsing (t),
When approximating C_jAs a decision method
Well known. Here, t represents time, and j is 0 to n
Is an integer. [0035] (Equation 1) The waveform signal y (t) handled this time is basically
Is a sine wave of frequency f,
Assume a sinusoidal function A · sin (2πft). And the waveform signal
The signal y (t) is generally represented by the sinusoidal function A · sin (2πf
t) includes a phase shift ψ, so FIG.
As shown, the waveform signal y (t) is converted into a sine wave component y (t) · sinψ
And the cosine wave component y (t) · cosψ. Accordingly
And the basis function φ_jAs (t), for the sine wave component:
Using Equations 2 and 3, and for the cosine wave component
The following equation 4 is adopted. [0037] (Equation 2) [0038] (Equation 3) [0039] (Equation 4) First, the waveform signal y (t) shown in FIG.
Time t_iSampling value y at _iAnd time t_iPair with
And collect m pairs of time-series data as
Basis function φ for sinusoidal component_jEach coefficient C of (t)_sin(C
_sin0, C_sin1, C_sin2…) And basis for cosine components
Function φ_jEach coefficient C of (t)_cos(C_cos0, C_cos1, C
_cos2…) Is calculated as follows. However,
Akira analyzes the components near frequency f in the waveform signal y (t).
It detects the amplitude, phase, etc.
No coefficient is needed for the next harmonic. Therefore,
In the case of, only the coefficients for n = 1 and n = 2 are examples
Show. The basis function for the sine wave component when n = 1
Number φ_jEach coefficient C of (t)_sin(C_sin0, C_sin1) And cosine wave
Basis function φ for components_jEach coefficient C of (t)_cos(C_cos0,
C_{co s1}) Is as shown in the following Expressions 5 and 6. [0042] (Equation 5) [0043] (Equation 6) Where P in the above formulas (5) and (6)_sin, Q_sin,
P_cos, Q_cosIs as shown in the following Expressions 7 to 10. What
Here, the number m of samples is represented by a basis function φ._jNext to harmonics of (t)
By making it sufficiently larger than the number n, the calculation accuracy can be improved.
Can be up. [0045] (Equation 7) [0046] (Equation 8) [0047] (Equation 9) [0048] (Equation 10) In the above description, the sampling interval and the sampling
The sampling number m was arbitrary, but in this case P^-1To
It has to be calculated every time, which is troublesome. Therefore,
For simplicity, the sampling frequency f_sIs assumed to be positive
The frequency f of the sinusoidal function A · sin (2πft) (that is,
Basis function φ_jFundamental wave of sine wave component and cosine wave component of (t)
An integer multiple of the frequency f), ie f_s= K_s・ F (k_sIs 3 or less
And the calculation cycle is the sine wave function
An integer multiple of the period of A · sin (2πft) (an integer of 1 or more)
k_tTimes), the sampling number m is m = k_t・ K_sNo
Swell. From these relationships and the periodicity of the trigonometric function,
The following Expressions 11 to 13 hold. [0050] (Equation 11) [0051] (Equation 12) [0052] (Equation 13) Then, the relationship of the above equations 11 to 13 is used.
And P in Equations 7 and 9_sin, P_cosIs the following number 14
Is represented as [0054] [Equation 14] Further, Q in the above equations (8) and (10)_sin, Q_cosTo
Sin (2πft)_i), Cos (2πft_iIn the calculation of)
Also t_iIn other words, the sampling period is
Refer to the constant table (sine wave function table) in
This can be done more easily. As a result, the calculation
By substituting the result into the above equations (5) and (6), the sine wave
Basis function φ for components_jEach coefficient C of (t)_sin(C_sin0,
C_sin1) And the basis function φ for the cosine wave component_jEach of (t)
Coefficient C_cos(C_cos0, C_cos1) Can be calculated easily
become. Next, the case where n = 2 will be described. This
, The basis function φ for the sinusoidal component_jEach coefficient of (t)
C_sin(C_sin0, C_sin1, C_sin2) And the cosine wave component
Basis function φ_jEach coefficient C of (t)_cos(C_cos0, C_cos1,
C_cos2) Is as shown in the following Expressions 15 and 16. [0057] (Equation 15) [0058] (Equation 16) Where P in Equations 15 and 16_sin, Q
_sin, P_cos, Q_cosIs as shown in the following Expressions 17 to 20.
You. In this case as well, the number m of samples is replaced by the basis function φ.
_jTo make it sufficiently larger than the harmonic order n of (t)
Thus, calculation accuracy can be improved. [0060] [Equation 17] [0061] (Equation 18) [0062] [Equation 19][0063] (Equation 20) Then, also in this case, f_s= K_s・ F (however
And this time k_sIs an even number of 4 or more) and m = k_t・ K_sconnection of
When using and applying the periodicity of the trigonometric function,
In addition to the relations of Equations 11 to 13, the following relation of Equation 21 is established.
I do. [0065] (Equation 21) Using these relationships, the above equations (17), (1)
9 P_sin, P_cosAre represented as shown in Equation 22 below.
And P_sin ^-1, P_cos ^-1Are as shown in Equation 23 below
To be represented. [0067] (Equation 22) [0068] (Equation 23) Further, Q in the above equations (18) and (20)_sin, Q_cos
Sin (2πft_i), Cos (2πft_i) Calculation
Even refer to the constant table (sine wave function table)
Therefore, the above-mentioned n = 1
Same as case. As a result, the calculation result P
_sin ^-1, P_cos ^-1, Q_sin, Q_cosInto the above equations 15 and 16
, The basis function φ for the sinusoidal component
_jEach coefficient C of (t)_sin(C _sin0, C_sin1, C_sin2) And the remainder
Basis function φ for sinusoidal component_jEach coefficient C of (t)_cos(C
_cos0, C_cos1, C_cos2) Can be easily calculated
You. B. Application example 1 Next, the basic theory as described above is applied to
Includes a sine wave measuring device for measuring the amplitude, phase, etc. of a sine wave
At the same time, various physical quantities are measured using the measurement of the sine wave.
The measuring device to be determined will be described. This measuring device, as shown in FIGS.
A microcomputer 10 is provided. Microcon
The computer 10 has a CPU, a ROM, a RAM, and the like,
Interface including A / D converter, D / A converter, etc.
The analog sine wave signal A₀・
sin2πft is output to the output terminal OUT and
Signal to be measured at input terminal IN (signal close to sine wave) y (t)
Enter For example, as shown in FIG.
Issue A₀・ Sin2πft is measured with the constant impedance element 21.
The series circuit 2 of the variable impedance element 22 to be fixed
0 is applied. In this case, the constant impedance element 21
The impedance of Z₀And variable impedance
Let the impedance of the dance element 22 be Zx. Soshi
And the constant impedance element 21 and the variable impedance element
The analog voltage signal divided by the element 22 is
The signal is input to the microcomputer 10 as a signal y (t).
You. For example, as shown in FIG.
Sine wave signal A₀· Sin2πft is a constant impedance element 3
1 to 33 and the variable impedance element 3 to be measured
4 is applied to a pair of diagonal positions of the bridge circuit 30 composed of
It is. In this case, the constant impedance elements 31 to 33
Impedance Z₁, Z_Two, Z_ThreeAnd the variable
The impedance of the impedance element 34 is defined as Zx.
Then, each of the other pair of diagonal positions of the bridge circuit 30 is turned on.
The differential voltage signal of the voltage
Is input to the microcomputer 10 as the number y (t).
You. In these cases, the variable impedance element 2
2, 34 correspond to various physical quantities such as temperature, force, displacement, etc.
Electrical element (e.g.,
Sensors for detecting the various physical quantities)
I have. The microcomputer 10 has the basic
Based on sine and / or cosine wave functions explained in principle
Function φ_jThe signal under test y using the least squares method as (t)
(t) is analyzed. This analysis is based on the basis function φ_j(t)
Each coefficient C_jIs performed by calculating
Number C_jAnd the reference sine wave signal A₀・ By comparison with sin2πft
The impedance of the variable impedance elements 22 and 34
Zx is detected. This change in impedance Zx
Since it changes depending on the various physical quantities,
The various physical quantities are measured from the analysis. Next, the signal under test y (t) is analyzed and
The impedance Zx of the variable impedance elements 22 and 34
Will be described in detail. This analysis and measurement
3 and 4, stored in the ROM of the microcomputer 10.
The executed program is executed by the CPU of the computer 10.
This is done by
FIG. 5 is a functional block showing functions realized by execution.
This will be described in detail with reference to the drawings. In this case, as shown in FIG.
Output from the microcomputer 10
, The analog reference sine wave signal Vout = A₀・ Sin
2πft is output, and the computer 10 has an input terminal
Sine wave signal Vin = Ax · sin (2πft +
It is assumed that a signal under measurement y (t) including ψ) is input. The microcomputer 10 has a reference clock.
And a timing signal generator 110.
ing. The reference clock generator 100 performs the analysis and measurement.
Generates a clock signal that serves as a fixed reference. Timing signal
The signal generator 110 receives the clock signal and
Various timing control signals that define the timing of operation
Is output. The microcomputer 10 has a
A sine wave generator 120 for generating a digital sine wave signal
ing. This sine wave generation unit 120
Multiple sampling data representing the instantaneous value of the sine wave
For multiple addresses corresponding to phases that increase by small angles
It has a sine wave table stored in response to
The timing control signal from the
Sampling data stored in the table
Digital sine wave signal by reading the
It is supplied to the D / A converter 122. D / A converter 122
Converts the supplied digital sine wave signal to D / A
Then, an analog reference sine wave signal is output to the output terminal OUT.
Vout = A₀・ Output as sin2πft. An A / D converter 130 is connected to the input terminal IN.
It is connected. The A / D converter 130 outputs the timing signal.
Control by the timing control signal from the signal generator 110
Then, the signal under test y (t) input to the input terminal IN is
Sampling at the specified rate and same sampling
A / D-converts the converted analog signal, and calculates a sine wave coefficient
140a and the cosine wave coefficient calculation unit 140b.
You. Note that the timing of the A / D conversion of the signal under test y (t) is
The D / A conversion section in the D / A conversion section 122.
It is preferable to match with the imming. The sine wave coefficient calculation unit 140a calculates the timing
Controlled by the timing control signal from the signal generator 110
The timing is controlled every m times the timing of the A / D conversion.
A / D conversion is performed using m sampling data.
Perform the operations of numbers 5, 7, 8 (or equations 15, 17, 18).
Go to the basis function φ for the sinusoidal component_jEach coefficient of (t)
C_sin(C_sin0, C_sin1) (Or each coefficient C)_sin(C
_sin0, C_sin1, C_sin2)). Cosine wave coefficient
The operation unit 140b also receives a signal from the timing signal generation unit 110.
The A / D conversion is controlled by a timing control signal.
For each m times the conversion timing, the m A / D converted m
Equations (6), (9), (10) using sampling data (or
By performing the operations of Equations 16, 19, and 20),Cosine waveIngredient
Basis function φ for_jEach coefficient C of (t)
_cos(C_cos0, C_cos1) (Or each coefficient C)
_cos(C_cos0, C_cos1, C_cos2))
Calculate. The performance of these sine wave coefficient calculation units 140a
The calculation and the calculation in the cosine wave coefficient calculation unit 140b are the same.
This is performed in synchronization with one sampling data. As described above, the input signal y (t)
Sampling frequency f_s(That is, the A / D conversion unit 130
Output cycle of A / D conversion) and sine wave coefficient calculation unit 140a
And the calculation cycle of the cosine wave coefficient calculation unit 140b is as follows:
, The coefficient calculation units 140a and 140b
Each coefficient C_sin(C_sin0, C_sin1), C_cos(C_cos0, C
_cos1) Can be simplified. In this case, the input signal
sampling frequency f of y (t)_sIs the standard assumed above.
Sine wave signal A₀The frequency f of sin (2πft) (ie,
The basis function φ_jbase of sine and cosine wave components of (t)
An integral multiple of the frequency f) of the main wave, ie, f_s= K_s・ F (just
Then k_sIs an integer of 3 or more).
Quasi-sine wave signal A₀A / D is the cycle 1 / f of sin (2πft)
Output cycle 1 / f of converter 130_sInteger multiple of 1
f = k_s・ 1 / f_sSet to. The sine wave coefficient calculation unit
140a and the cosine wave coefficient calculation unit 140b
Sine wave signal A₀-An integral multiple of the cycle of sin (2πft) (1 or less)
The integer k above_tTimes). As a result, each coefficient
Calculation unit 140a, 140b each coefficient C_sin(C_sin0, C
_sin1), C_cos(C_cos0, C_cos1) To calculate
The sampling number m used is m = k_t・ K_sBecomes Therefore, also in this case,
13 holds, and each of the coefficients C_sin(C_sin0, C_sin1),
C_cos(C_cos0, C_cos1) To calculate the numbers 5 and 6
P_sin, P_cosIs calculated by the above equation (14). this
So that the sampling frequency f_s(That is, A / D conversion
Output cycle of A / D conversion of unit 130) and sine wave coefficient calculation
Calculation cycle of the unit 140a and the cosine wave coefficient calculation unit 140b
Of the reference sine wave signal A
₀・ Since sin (2πft) is known, it can be easily achieved.
And The sine wave coefficient calculating section 140a and the cosine
Each coefficient C by the wave coefficient calculation unit 140b_sin(C_sin0, C
_sin1, C_sin2), C_cos(C_cos0, C_cos1, C_cos2)of
In order to simplify the arithmetic processing, the above conditions must be added.
And the coefficient k_sMust be set to an even number greater than or equal to 4.
You. According to this, as described above,
In addition, Equation 21 is established, and each of the coefficients C_sin(C_sin0, C
_sin1, C_sin2), C_cos(C_cos0, C_cos1, C_cos2)
P of numbers 15 and 16 for calculation_sin ^-1, P_cos ^-1Is on
It is calculated by the notation 23. In this way, each of the coefficient calculation units 140a, 1
Each coefficient C calculated in 40b_sin(C_sin0, C_sin1)
(Or each coefficient C_sin(C_sin0, C_sin1, C_sin2) And each
Coefficient C_cos(C_cos0, C_cos1) (Or each coefficient C)_cos(C
_cos0, C_cos1, C_cos2)), Each coefficient C_sin1, C
_cos1Are supplied to the amplitude calculation section 150a and the phase calculation section 150b.
Supplied respectively. The amplitude calculation unit 150a calculates
Calculation (coefficient C_sin1, C _cos1Of the square root of the sum of each square of
Calculation), the sine wave signal Vin = Ax · sin (2
The amplitude Ax of (πft + π) is calculated. Phase calculator 150b
Is calculated by the following equation 25 (coefficient C_sin1, C_cos1Of the value of the ratio of
Execution of the arctangent operation)
Wave signal Vin = Ax · sin (2πft + ψ) Phase shift amountψ
Is calculated. Note that the amplitude is expressed by
The calculation of Ax and the phase shift amount ψ is based on each value C
_sin1, C_cos1Is a sine wave signal Vin = Ax · sin (2πft +
Basis function φ corresponding to the frequency of ψ)₁Sine wave as (t)
The coefficients of the component sin2πft and the cosine wave component cos2πft
It is clear from the fact that there is and that the vector diagram in FIG.
Is. [0084] [Equation 24] [0085] (Equation 25) The amplitude calculator 150a and the phase calculator 150
the amplitude Ax and the phase shift calculated respectively in b
ψ is supplied to the variable impedance calculator 160.
The variable impedance calculator 160 calculates the amplitude Ax and
Variable impedance element 2 based on phase shift amount ψ
Calculate impedance Zx of 2,34 (FIGS. 3 and 4)
You. In this calculation, the output terminal OUT is
The base supplied to the column circuit 20 and the bridge circuit 30 of FIG.
Quasi-sine wave signal Vout = A₀・ Sin2πft is already known.
3 and 4 constant impedance elements 21 and 31 to 33
Peedance Z₀, Z₁, Z_Two, Z_ThreeAre also known. And
Amplitude A₀, Ax, the amount of phase shift ψ, and each impedance Z
₀, Z₁, Z_Two, Z_ThreeIs related to the case of the series circuit 20 in FIG.
In this case, the following equations 26 and 27 hold, and
In the case of the bridge circuit 30, the following equations 28 and 29 hold.
I do. However, in the following equations 26 to 29, Z₀, Z₁,
Z_Two, Z_Three, Zx represents a complex number, and Re represents the number in parentheses.
Represents the real part of the complex number, and Im represents the complex number in parentheses.
Represents the imaginary part. [0087] (Equation 26) [0088] [Equation 27] [0089] [Equation 28] [0090] (Equation 29) Then, these equations 26, 27 or 2
8 and 29, the amplitude calculator 150a and the phase calculator 15
Substituting the amplitude Ax and the phase shift amount 計算 calculated at 0b
If other impedance Z₀~ Z_ThreeIs known,
Each impedance Z of the variable impedance elements 22 and 34
x = Rx + jXx can be calculated. Paraphrase
For example, the resistance Rx of the impedance Zx and the reactance X
x can be calculated and the impedance Zx
(Absolute value) | Zx | = (Rx^Two+ Xx^Two)^1/2And Inn
Phase shift tan due to impedance Zx^-1(Xx / Rx)
Can calculate. And these variable impedance elements
The impedance Zx of 22 and 34 is temperature, force, displacement
Which is configured to change according to various physical quantities
For example, various physical quantities such as temperature, force, and displacement can be measured. As described above, according to the application example 1, the input
M sampling data obtained by sampling the force signal y (t)
Applying the least squares method to the sine wave coefficient calculation unit 140
a and the reference sine wave signal V in the cosine wave coefficient calculation unit 140b.
out = A₀A sinusoidal function with the same frequency as sin2πft and the remainder
Each coefficient C of the sine wave function_sin1, C_cos1Was calculated respectively.
Then, the amplitude calculator 150a, the phase calculator 150b,
The variable impedance calculator 160 calculates the coefficient
C_sin1, C_cos1Using the variable impedance element 22,
34, each impedance Zx = Rx + jXx, impedance
Depends on the magnitude | Zx | of the impedance Zx and the impedance Zx.
Phase shift amount tan^-1(Xx / Rx) was calculated. I
Therefore, according to this, it corresponds to the sampling frequency.
Higher than the accuracy specified by the A / D converter 130.
The impedance Zx = Rx + jXx in degrees and the phase shift
Quantity tan^-1(Xx / Rx) can be calculated. Further, each of the coefficient calculation units 140a, 140
Each coefficient C calculated in b_sin(C_{s in0}, C_sin1)(or
Is the coefficient C_sin(C_sin0, C_sin1, C_sin2)) And each person in charge
Number C_{co s}(C_cos0, C_cos1) (Or each coefficient C)
_cos(C_cos0, C_cos1, C_cos2)) Out of the rest
Number C_sin0, C_cos0(Or coefficient C_sin0, C_sin2, C_cos0,
C_cos2) Are supplied to the abnormality detection unit 170. Abnormality detector
170 is the value of these supplied coefficients C _sin0, C_cos0(or
Is the coefficient C_sin0, C_sin2, C_cos0, C_cos2)On the basis of the,
Judgment of abnormalities in impedance measurement. Coefficient C_sin0,
C_cos0Is the reference sine wave signal Vout = A₀・ Sin2πft
DC offset (DC noise component).
Also, the coefficient C_sin2, C_cos2Is the reference sine wave signal Vout =
A₀・ Harmonic component to sin2πft (harmonic noise component
Minutes). Therefore, the sum of each square of these coefficients is
Square root of (C_sin0 ^Two+ C_cos0 ^Two)^1/2, (C_sin2 ^Two+ C_cos2 ^Two)
^1/2By calculating, the DC off unnecessary for the measurement
The size of the set and harmonic components can be measured. And this
Measurement of these DC offsets, the magnitude of harmonic components, etc.
By comparing the value with a predetermined value, the impedance
Detects measurement abnormalities and generates a detection signal indicating the abnormalities.
Can output. In the above application example 1, the variable input
The series circuit 20 having the impedance elements 22 and 34 or the
Each impedance of the ridge circuit 30 corresponds to a resistance component and a reactor.
General examples that can be applied even if
Was. However, each of the variable impedance elements 22 and 34
The impedance Zx is equal to the resistance Rx or the reactance Xx.
If there is only one, the microcomputer 10
Calculations can be simplified. However, in this case
Are the impedances of the constant impedance elements 21 and 31 to 33.
-Dance Z₀, Z₁, Z_Two, Z_ThreeNor resistance or reactor
Only one of the components. As described above, the resistance circuit 20 and the bridge circuit
The path 30 must contain only resistance or reactance
Is included in the input signal y (t) input to the input terminal IN.
The component of the frequency f is output from the output terminal OUT.
Reference sine wave signal Vout = A₀・ Same phase as sin2πft
, The sum of the cosine wave components of the input signal y (t)
Calculation is not required. The phase shift amount ψ is also “0”.
Therefore, the phase calculation unit 150b is not required. Therefore,
In this case, the functional block diagram of FIG. 5 is written as shown in FIG.
Be replaced. In the functional block diagram of FIG.
Quasi-clock generator 100, timing signal generator 11
0, sine wave generator 120, D / A converter 122, A / D
The conversion unit 130 and the sine wave coefficient calculation unit 140a
The operation is the same as in the case of the functional block diagram of FIG. Amplitude calculation
The section 150a is calculated by the sine wave coefficient calculating section 140a.
Coefficient C_sin1Input only and output as amplitude Ax
I do. This means that the cosine wave component of the input signal y (t) is “0”
And the coefficient C_sin1Is the reference sine wave of the input signal y (t)
Signal Vout = A₀・ Frequency corresponding to frequency f of sin2πft
This is because it is equal to the amplitude Ax of several components. In other words,
The amplitude calculation section 150a becomes unnecessary. The variable impedance calculator 160
Is a variable impedance based on the above equations (26) and (28).
Impedances Xx of the elements 22 and 34 (FIGS. 3 and 4).
That is, either the resistance component Rx or the reactance component Xx
Is calculated. This also allows the variable impedance element
22 and 34 have only resistance Rx or reactance Xx
And the same resistance Rx or reactance Xx
It changes according to various physical quantities such as degree, force, displacement, etc.
If so, measure various physical quantities such as temperature, force, displacement, etc.
it can. Further, the abnormality detecting section 170 is provided in the same manner as described above.
Coefficient C supplied_sin0(Or C_sin0, C_sin2Based on)
As in the case of the functional block diagram of FIG.
Measures the magnitude of unnecessary DC offset and harmonic components
You. Therefore, in this case as well, these DC offsets
Comparison of measured values such as the magnitude of the
To detect impedance measurement abnormalities
At the same time, a detection signal indicating the abnormality can be output. As described above, input to the input terminal IN
The component of the frequency f included in the input signal y (t) is
Reference sine wave signal Vout = A output from input terminal OUT₀
-Even if it is the same as sin2πft, the function block shown in FIG.
Cosine described above in place of the sine wave coefficient calculation unit 140a in FIG.
It is also possible to use the wave coefficient calculator 140b.
You. In this case, under the control of the timing signal generator 110,
A / D conversion timing in the A / D conversion unit 130
Calculation timing in the cosine wave coefficient calculation unit 140b
The cosine wave coefficient calculation unit 140b controls
The phase of the signal y (t) is set to the reference sine wave signal Vout = A₀・ Sin2π
If we calculate by shifting ft / 2 by π / 2,
Good. In this way, the variable impedance element
The impedance Zx of 22 and 34 is the resistance Rx or reactor
In addition to the resistance Xx, the resistance Rx or the rear
Xx is converted into various physical quantities such as temperature, force, displacement, etc.
Temperature, force, displacement, etc.
Various physical quantities can be measured. In this case, the abnormality detector 170
Is the coefficient C supplied from the cosine wave coefficient calculator 140b.
_cos0(Or C_cos0, C_cos2) Based on the machine shown in FIG.
As in the case of the functional block diagram, DC offset
And the magnitude of harmonic components. Therefore,
In the case of
By comparing a measured value such as
Detects abnormalities in impedance measurement and displays the abnormalities.
Output a detection signal. Further, in the above application example 1, the variable input
Phase shift due to the impedance elements 22 and 34 (FIGS. 3 and 4)
Independent of the magnitude | Zx |
Is measured, the phase of the input signal y (t) is used as the reference sine.
Wave signal Vout = A₀・ Forcibly adjust to sin2πft and enter
Even if only the sine wave component of the force signal y (t) is calculated,
Good. Specifically, as shown in FIG.
The zero cross detection unit 180 is added to the block diagram to
The timing signal generating unit 11 is
0 is controlled. The zero-cross detecting section 180 receives the input signal y
Timing that (t) changes from negative to positive (or from positive to negative)
Is detected from the output terminal OUT.
Reference sine wave signal Vout = A₀・ Input for sin2πft
The phase shift amount of the signal y (t) is calculated. And Zeroku
The loss detector 180 calculates the phase shift amount based on the calculated phase shift amount.
The timing signal generator 110 is controlled, and the A / D converter is controlled.
130: A / D conversion timing and sine wave coefficient operation
The operation timing in the arithmetic unit 140a is controlled. this
In this case, the input signal y (t) is the reference sine wave signal Vout = A₀・ Sin
Since the waveform is almost the same as 2πft, the sine wave coefficient calculation
The calculation cycle of the input signal y (t) in the unit 140a is
Sine wave signal Vout = A₀・ Synchronizing with sin2πft
Wear. As a result, in this case, the reference sine wave signal Vout
= A₀・ Analyze input signal y (t) with the same phase as sin2πft
Will be. This analysis is based on the functional block diagram of FIG.
This is the same as the analysis, and the amplitude calculation unit 1
50a is a coefficient calculated by the sine wave coefficient calculation unit 140
C_{si n1}Is the amplitude Ax and the variable impedance calculator 16
Output to 0. However, the coefficient C in this case_sin1And amplitude
In the calculation of Ax, the input signal y (t) and the reference sine wave signal
No. Vout = A₀・ Ignore the phase shift with sin2πft and enter
Sine wave component C included in force signal y (t)_sin1・ Sin2πft
With reference sine wave signal Vout = A₀・ Forcibly same as sin2πft
Ax is a variable impedance element
Series circuit 20 or variable impedance element 3 including
4 to the magnitude of the impedance of the bridge circuit 30
Depends on the value. Also in this case, the above equations (26) and (28) are
It holds. Therefore, the variable impedance calculator
160 is a variable input by solving the above equations 26 and 28.
Of the impedance Zx of the impedance elements 22 and 34
| Zx | can be calculated. However, in the above equations (26) and (28), a
Impedance Zx is other impedance Z₀, Z₁, Z_Two,
Z_ThreeIs expressed in a format that includes
Unless added, the magnitude | Zx |
It cannot be taken out alone. However,
As in the modification of the application example 1, the variable impedance element 2
2,34 can ignore the reactance component or reduce the resistance component
If it can be ignored, other impedances Z₀,
Z₁, Z_Two, Z_ThreeCan also ignore reactance or resistance
By using the above, the above equations 26 and 28 are
It is deformed like 30, 31. [0108] [Equation 30] [0109] [Equation 31] Therefore, according to this modification, the variable
If the impedance elements 22 and 34 are resistance Rx or
And has the same resistance Rx or
Actance Xx can be used for various physical quantities such as temperature, force, displacement, etc.
Temperature, force, displacement, etc.
Can measure various physical quantities. Also in this modified example, the abnormality detecting section
170, as in the case of the functional block diagram of FIG.
An abnormality in impedance measurement is detected. In addition,
In the modification of the above, the control of the zero-cross detection unit 180
From the calculation cycle in the sine wave coefficient calculation unit 140a,
Quasi-sine wave signal Vout = A₀・ Synchronized with the same phase as sin2πft
I tried to make it. However, instead of this,
The sine wave coefficient calculating unit 140a is controlled by the detecting unit 180.
, The reference sine wave signal Vout = A₀・ Sin2
πft or π / 2 or -π / 2
You may. In this case, the sine wave function shown in the functional block diagram of FIG.
Cosine wave coefficient calculation unit 1 described above in place of number calculation unit 140a
40b, the amplitude calculator 150a calculates the cosine wave coefficient
Coefficient C calculated by the unit 140b_cos1Enter only
Thus, it is preferable to output the amplitude Ax. Further, the above-mentioned application example 1 and various modifications thereof
In, the sine wave coefficient calculation unit 140a and the cosine wave coefficient
Calculation unit 140b, amplitude calculation unit 150a, phase calculation unit 150
b, variable impedance calculation section 160 and abnormality detection section 1
In 70, every m sampling data,
Reference analog sine wave included in the input analog signal
Of the sine wave component and / or cosine wave component of the same frequency as the signal
Period k_tTimes (k_tIs an integer greater than or equal to 1)
To be executed. However, the input analog signal
And the reference sine wave signal Vout = A₀・ Sin2πft
Sine wave component and / or cosine wave component of the same frequency f
For each cycle, various types using the m sampling data
The calculation may be performed. Note that k_tIs "1"
If it is larger than the above, the execution of the operation
Minute sine wave component and / or cosine wave component
The calculation is performed sequentially while shifting. In addition,
In addition, one or more sampling
Use the m sampling data at each timing
Various calculations may be performed. C. Application example 2 Next, the basic theory described above is applied to the rotation angle of the rotor with respect to the stator.
θ (specifically, the electrical
Angle θ) to measure the resolver of an electric motor
Application Example 2 will be described. FIG. 9 schematically shows the measuring device.
The winding (coil) connected to the microcomputer 10
) 41-43. Winding 41 is an electric motor
The winding 41 is provided on the stator side, and the measuring device 10
Reference sine wave signal A via output terminal OUT of₀・ Sin2π
ft is applied. Windings 42 and 43 are electric motors
Is provided on the rotor side of the
The voltage induced by the rotation of line 41 is measured as signal y
₁(t), y_Two(t) via the input terminals IN1 and IN2
It is supplied to the microcomputer 10. The winding 41
Are provided on the rotor side, and the windings 42 and 43 are
It may be provided on the side. If the windings 42 and 43 are configured identically,
They are located at electrical angles shifted by π / 2 from each other
And the electrical angle of the rotor winding with respect to the stator winding is θ.
Then, the signal under measurement y₁(t), y₂(t) is Ax1 · A₀・ Sin
θ · sin (2πft + ψ), Ax1 · A₀・ Cosθ ・ sin (2πft
+ Ψ). That is, DC drift,
If unnecessary components such as harmonic noise are not included, y₁(t)
= Ax1 · A₀・ Sinθ ・ sin (2πft + ψ), y_Two(t) = Ax2 ·
A₀Cos θ · sin (2πft + ψ) Note that ψ indicates the base
Quasi-sine wave signal A₀· Sin2πft and the signal under measurement y₁(t), y
_Two(t) is the same as that described above indicating the amount of phase shift.
You. Also, the windings 42 and 43 have the same configuration.
In an ideal state with no abnormalities, the amplitude Ax1 · A₀, Ax2
・ A₀Have the same value. Next, as in the case of the application example 1, the personal computer
The operation of the microcomputer 10 is shown in the functional block diagram of FIG.
This will be described with reference to FIG. In this case, too,
The reference clock generator 100 of the application example 1
Timing signal generator 110, sine wave generator 120,
D / A conversion unit 122 and A / D conversion unit 130, respectively
Reference clock generator 200, which functions similarly, timing
Signal generator 210, sine wave generator 220, and D / A conversion
And an A / D converter 130.
A / D converters 230 and 240 functioning similarly are provided.
You. The A / D converter 230 is connected to the input terminals IN1 and IN1.
Connected to the timing signal generator 210,
Controlled by timing control signal from signal generator 210
The signal under test y input at the input terminal IN1
₁(t) Sampling (SIN phase signal) at a predetermined rate
A / D conversion of the sampled analog signal
In other words, the sine wave coefficient calculation unit 250a and the cosine wave coefficient calculation
Output to the unit 250b. The A / D converter 240 has an input terminal
Child IN2 and the timing signal generator 210,
Timing control signal from timing signal generator 210
Measured by the input terminal IN2
Signal y_Two(t) Sampling (COS phase signal) at a predetermined rate
And the analog signal sampled at the same
/ D conversion, the sine wave coefficient calculating section 260a and the cosine wave
It outputs to the number operation part 260b. A / D converters 230 and 240 are provided.
Instead, the SIN phase signal and the COS phase signal
One A / D conversion unit that performs A / D conversion as described above
The time division A / D converted SI
Distributes N-phase digital signal and COS-phase digital signal
You may force it. The sine wave coefficient calculation units 250a and 260a
It functions in the same manner as the sine wave coefficient calculation unit 140a,
The above using m / D converted m sampling data
Performing the operations of Equation 5, Equation 7 (or Equation 14) and Equation 8,
Basis function φ for sine wave component_jCoefficient C of (t)_sin1To
Calculate each. Cosine wave coefficient calculator250b, 260
bAlso functions in the same manner as the cosine wave coefficient calculation unit 140b,
Using the A / D converted m sampling data,
Perform the operations of Equations 6, 9 (or 14) and 10
And the basis function φ for the cosine wave component_jCoefficient C of (t)
_cos1Is calculated respectively. In these cases,
Has the same frequency as the reference sine wave signal Ao · sin2πft.
Each coefficient C of sine wave component and cosine wave component_sin1, C
_cos1Only calculate each. Sine wave coefficient calculation section 250a and cosine wave coefficient
Coefficient C calculated by operation unit 250b_sin1, C_cos1Is
The amplitude is supplied to the amplitude calculator 270. Sine wave coefficient calculator 26
0a and the coefficient calculated by the cosine wave coefficient calculation unit 260b
C_sin1, C_cos1Is supplied to the amplitude calculator 270. Shake
The width calculators 260 and 270 respectively include the amplitude calculator 150
a, the reference sine wave signal A₀・ Sin2πft
Measured signal y of the SIN phase having the same frequency as₁(t) = Ax1 · A
₀・ Sinθ ・ sin (2πft + ψ) and COS phase signal to be measured
y₂(t) = Ax2 · A₀・ Each amplitude of cosθ ・ sin (2πft + ψ)
Asin = Ax1 · A ₀・ Sin θ, Acos = Ax2 ・ A₀・ Cosθ
Calculate each. Each of these amplitudes Asin = Ax1 · A₀・ Sinθ,
Acos = Ax2 ・ A₀Cosθ is the rotation of the rotor with respect to the stator
The turning angle θ (electrical angle θ of the rotor winding with respect to the stator winding)
To the rotation angle calculation unit (electric angle calculation unit) 300 for calculation
Supplied. The windings 42 and 43 are configured identically.
And if no abnormality has occurred, the values Ax1 and Ax2 are
equal. The rotation angle calculator (electrical angle calculator) 300 is as follows:
The rotation angle θ (electrical angle θ) is calculated by executing the calculation of Expression 32.
Calculate. [0122] (Equation 32) According to this, the same as in the case of the above-mentioned application example 1 is obtained.
In addition, the rotation angle θ (electrical angle θ) of the electric motor
₁(t), y_Twom samplings sampled from (t)
Sinusoidal components calculated by applying the least squares method to the data and
Each coefficient C of cosine wave component_sin1, C_cos1Is calculated based on
You. Therefore, the A / D corresponding to the sampling frequency
Higher accuracy than the accuracy specified by the converters 230 and 240
The rotation angle θ (electrical angle θ) can be calculated in degrees. In Equation 32, the value of the denominator is
The amplitude values Asin and Acos become negative values.
In such cases, the calculation of Equation 32 becomes troublesome. did
Accordingly, as shown in the following Expressions 33 to 40, the amplitude value Asi
Different operations depending on the values of n and Acos (however,
4 is an operation equivalent to Equation 32)
The calculation of the rotation angle θ can be easily performed. Under these
In notations 33 to 44, the amplitude values Asin and Acos are
Let a = Asin and b = Acos, respectively. Where a = A
Calculation of rotation angle θ when sin = 0 and b = Acos = 0
Is impossible and the values a and b are abnormal values,
Do not calculate the rotation angle θ. [0125] [Equation 33] [0126] [Equation 34] [0127] (Equation 35) [0128] [Equation 36] [0129] (37) [0130] [Equation 38] [0131] [Equation 39] [0132] (Equation 40)Also, the amplitude calculation units 260 and 270 calculate
Each amplitude Asin = Ax1 · A₀・ Sin θ, Acos = Ax2 ・ A₀
・ Measured by cosθ and rotation angle calculator (electric angle calculator) 300
The calculated rotation angle θ (electrical angle θ) is sent to the abnormality detection unit 310.
Is also supplied. The abnormality detection unit 310 calculates
42, the SIN phase amplitude value Ax1 · A₀Passing
And COS amplitude value Ax2 · A₀Is calculated respectively. However
However, in Equations 41 and 42 below, the denominator sin θ and
If cosθ becomes a value near “0”, the calculation becomes impossible.
Where the absolute values of sinθ and cosθ | sinθ | and | cosθ |
If the value is smaller than a certain small value, the following Expression 41 and Expression 4
2 is not performed. [0134] (Equation 41) [0135] (Equation 42) Then, the abnormality detecting section 310
Width value Ax1 ・ A₀, Ax2 ・ A₀Is within a predetermined range
Therefore, an abnormality in the measurement of the rotation angle θ (electrical angle θ) is detected.
This causes the windings 41 to 43 to be short-circuited internally.
Abnormalities such as rare short, sine wave applied to winding 41
If the SIN phase and COS phase are distorted,
Amplitude value Ax1 · A₀, Ax2 ・ A₀Is an abnormal value
Have been. Therefore, according to the application example 2, the rotation angle
Abnormal measurement of θ (electric angle θ) is the same rotation angle θ (electric angle θ)
It is easily detected in combination with the measurement of In addition, this
In the example of the above, both amplitude values A
x1 ・ A₀, Ax2 ・ A₀Is calculated individually,
The amplitude value of only one of them was calculated and calculated
The abnormality may be determined only by the amplitude value. In this application example 2, the SIN phase
Measured signal y of₁The measured signal of the COS phase is also
No. y_TwoAlso for (t), the sine wave coefficient calculation unit 25
0a, 260a and cosine wave coefficient calculation units 250b, 260
b is provided respectively. However, the windings 42, 4
3 is the reference sine wave applied to the winding 41
Signal A₀・ The phase is shifted by approximately π / 2 from sin2πft
Signal. Therefore, the timing signal generator 210
Output terminal OUT by the control signal output from
Reference sine wave signal A₀・ Same timing as sin2πft
In other words, the reference sine wave signal A₀・ Output of sin2πft
Each coefficient operation unit 250a, 250
50b, 260a, 260b
If the signal under test y₁(t), y_TwoThe sine wave component of (t) is theoretical
Should be "0". Therefore, the microcomputer 10
The calculation can be simplified as shown in FIG. Sand
That is, the sine wave coefficient calculation units 250a and 260a are omitted.
Wear. Also, the cosine wave coefficient calculation units 250b and 260b
Each coefficient C calculated by_cos1Are the SIN phase and
Each measured signal y of the COS phase₁(t), y_TwoStandard in (t)
Sine wave signal A₀· Amplitude Ax of the same frequency component as sin2πft
1 ・ A₀・ Sinθ 、 Ax2 ・ A ₀・ Because each represents cosθ,
The amplitude calculation units 270 and 280 are not required. Accordingly
The cosine wave coefficient calculation units 250b and 260b
Each calculated coefficient C_cos1Is the amplitude Ax1 · A₀・ Sin θ, Ax2
・ A₀A rotation angle calculator 3 as a signal representing cos θ;
00 and the abnormality detection unit 310. And rotation
As in the above case, the angle calculator 300 calculates
Calculates the rotation angle θ (electrical angle θ). Also abnormal
The detection unit 310, as in the above case,
Calculation of the above equations 41 and 42 also using the turning angle θ (electrical angle θ)
Ax1 · A₀, Ax2 ・ A₀And the rotation angle θ
An abnormality in the measurement of (electrical angle θ) is detected. The result
As a result, according to this, the rotation angle θ (electrical angle θ) can be easily measured.
As well as detect abnormalities in the measurement
Become like Further, the same as in the case of the modification of the above-mentioned application example 1
Next, as shown in FIG.
Then, the measured signal y of the SIN phase₁(t) = Ax1 · A₀・ Sinθ ・
sin (2πft + ψ) and the signal under test y in the COS phase_Two(t) =
Ax2 ・ A₀・ Phase shift at cos θ ・ sin (2πft + ψ)
It is also possible to set the quantity ψ to “0”. According to this,
The cosine wave coefficient calculation units 250b and 260b of the application example 2 are
Can be omitted. In this case, the sine wave coefficient calculation unit 2
50a, each coefficient C calculated by 260a_sin1But,
Each signal under measurement y of SIN phase and COS phase respectively
₁(t), y_TwoThe reference sine wave signal A in (t)₀・ Sin2πf
Amplitude Ax1 · A of the same frequency component as t₀・ Sinθ 、 Ax2 ・ A₀・ C
osθ. So in this case
In addition, the above-mentioned amplitude calculation units 270 and 280 are also unnecessary,
As in the above case, the angle calculator 300 calculates
The rotation angle θ (electrical angle θ) is calculated by
When the rotation angle θ (electrical angle θ) is measured by the normal detection unit 310,
Abnormality is detected. As a result, this also
When it becomes possible to simply measure the rotation angle θ (electrical angle θ)
In addition, the abnormality of the measurement can be detected. In the above application example 2 and various modifications thereof,
Are sine wave coefficient calculation units 250a and 260a, and cosine wave coefficient
Calculation units 250b and 260b, amplitude calculation units 270 and 28
0, in the rotation angle calculation unit 300 and the abnormality detection unit 310
Is every m sampling data, that is,
Same frequency as the reference analog sine wave signal included in the analog signal
K of period of sine wave component and / or cosine wave component of wave number_tDouble
(K_tIs an integer greater than or equal to "1").
I did it. However, in this case, too,
Signal with the same frequency as the reference analog sine wave signal
For each cycle of the sine wave component and / or cosine wave component, the m
Various operations using the sampling data of
You may. Note that k_tIs greater than "1",
By performing the various calculations, the sine wave for a plurality of cycles
Component and / or cosine wave component sequentially shifted by one period
Will be calculated. In addition, the calculation speed has been further increased
At one or more sampling timings,
Performs various calculations using the m pieces of sampling data
You may do so. D. Other Modifications In the functional block diagram of FIG.
The wave coefficient calculation unit 140a and the cosine wave coefficient calculation unit 140b
And always calculate both coefficients Csin and Ccos
The calculation unit 150b uses both coefficients Csin and Ccos to calculate the phase shift.
The amount れ was always calculated. But its variant
If only the amplitude Ax needs to be calculated as in
At the start of the calculation, the phase shift amount ψ is calculated at predetermined time intervals, etc.
Then, the sine wave coefficient calculation unit is
The calculation cycle of the input signal y (t) at 140a is defined as a reference sine.
Wave signal Vout = A₀・ Can be synchronized to sin2πft
You. In this case, based on the calculated phase shift amount ψ,
By controlling the timing signal generator 110, A
A / D conversion timing and correct
Controls the calculation timing in the sine wave coefficient calculation unit 140a
Then, the operation cycle of the input signal y (t) is set to the reference sine wave signal Vou.
t = A₀・ Synchronization with sin2πft is sufficient. According to this, the cosine wave coefficient calculating section 140b
And only occasionally perform the operation of the phase operation unit 150b,
Always use a sine wave as in the functional block diagram of FIG.
Calculating only the coefficient Csin in the coefficient calculation unit 140a
As a result, the amplitude Ax can be calculated. This
Microcomputer 10 in calculating the amplitude Ax.
Load can be reduced. In addition, the phase shift amount ψ
Occasional synchronization control allows time-dependent
As the temperature changes, the phase of the input signal y (t) changes.
Can always keep the calculation accuracy of the amplitude Ax good.
You. Also, in this case, the calculated phase shift
Control the timing signal generator 110 using the
The calculation cycle in the cosine wave coefficient calculation unit 140b is
Quasi-sine wave signal Vout = A₀・ Π / 2 or for sin2πft
May be shifted by -π / 2. This
For example, the sine wave coefficient calculator 140a and the phase calculator 150b
The cosine wave coefficient calculation is always
By calculating only the coefficient Ccos in the unit 140b,
The amplitude Ax can be calculated. Further, using such a phase shift amount ψ,
The control of the operation timing is different from the application example 2 shown in FIG.
Also applies. In this case, the amplitude calculation units 270 and 280
, The sine wave coefficient calculation units 250a and 260a and the cosine wave
Coefficients Csin, Cc from coefficient calculation units 250b, 260b
The calculation of the phase shift amount ψ based on os
Calculated from time to time and the calculated phase shift
Using タ イ ミ ン グ to control timing signal generator 210
Sine wave coefficient calculation units 250a and 260a and cosine
Input signal y in wave coefficient calculators 250b and 260b
The calculation cycle of (t) is defined as a reference sine wave signal Vout = A₀・ Sin2πf
t or the reference sine wave signal Vout = A₀・ Sin2πf
By shifting by t by π / 2 or -π / 2
Good. Thus, the total amount of the phase shift ψ can be calculated.
Calculation is performed only occasionally, and the sine wave coefficient calculation unit 2 is always used.
50a, 260a or cosine wave coefficient calculation units 250b, 26
Only calculate one of the coefficients Csin and Ccos at 0b
And the respective amplitudes Ax1 · A of the SIN phase and the COS phase₀・ Sinθ, A
x2 ・ A₀・ Cos θ can be calculated. And these
Ax1 · A₀・ Sinθ 、 Ax2 ・ A₀・ Rotation angle using cosθ
The calculation unit 300 calculates the rotation angle θ. As a result,
As a result, the microcompilation in the calculation of the rotation angle θ
The load on the computer 10 can be reduced. Also, the amount of phase shift ψ
By performing the above-mentioned synchronization control from time to time,
As the temperature or the like changes with the passage of time, the input signals y1 (t), y1
Even if the phase of (t) changes, the calculation accuracy of the rotation angle θ is always good.
You can keep it good. The above-mentioned application examples 1 and 2 and their modifications
In the example, the sine wave generators 120 and 220 and the D / A
A sine wave signal is generated by the converters 122 and 222.
Instead of generating a sinusoidal signal
Various electrical circuits are available. For example, an analog sine wave
An oscillator that generates a signal can be used. Further, the above-mentioned application examples 1 and 2 and their respective
In the variant, various operations are performed by a microcomputer.
10 but the digital signal processing
One or more hardware circuits (eg, integrated circuits)
), The functions of the microcomputer 10 are realized.
You may make it appear.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a waveform diagram of a sine wave signal which is a signal under measurement. FIG. 2 is a vector diagram for explaining a sine wave component and a cosine wave component of an input signal. FIG. 3 is an overall schematic diagram of a measuring apparatus according to an application example 1 to which the present invention is applied. FIG. 4 is an overall schematic diagram showing another example of the measuring device according to Application Example 1 to which the present invention is applied. FIG. 5 is a functional block diagram of the microcomputer shown in FIGS. FIG. 6A is a waveform diagram of an output signal and an input signal from the microcomputer, and FIG. 6B is a vector diagram of both signals. FIG. 7 is a functional block diagram showing a modification of the functional blocks of FIG. 5; FIG. 8 is a functional block diagram showing another modified example of the functional blocks of FIG. 5; FIG. 9 is an overall schematic diagram of a measuring device according to an application example 2 to which the present invention is applied. FIG. 10 is a functional block diagram of the microcomputer of FIG. 9; FIG. 11 is a functional block diagram showing a modification of the functional blocks shown in FIG. 10; FIG. 12 is a functional block diagram showing another modified example of the functional blocks shown in FIG. 10; [Description of References] 10 ... microcomputer, 20 ... series circuit, 21 ...
Constant impedance element, 22: variable impedance element, 30: bridge circuit, 31, 32, 33: constant impedance element, 34: variable impedance element, 41 to 41
43 ... winding, 110, 210 ... timing signal generator,
120, 220: sine wave generator, 130, 230, 24
0: A / D converter, 140a, 250a, 260a: sine wave coefficient calculator, 140b, 250b, 260b: cosine wave coefficient calculator, 150a, 270, 280: amplitude calculator, 150b: phase calculator, 160 ... Variable impedance calculation units, 170, 310 ... abnormality detection units, 180, 32
0: Zero cross detection unit; 300: Rotation angle calculation unit.

Continuation of front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G01B 7/30 G01D 5/245

Claims (1)

(57) [Claim 1] A first winding is provided on one of a stator and a rotor, and the other of the stator and the rotor is opposed to the first winding and mutually opposed. In a rotation angle detecting device provided with orthogonal second and third windings to detect a rotation angle of a rotor with respect to a stator, reference signal applying means for applying a predetermined reference analog sine wave signal to the first winding And inputting an induced voltage signal from the second winding as a first measurement analog signal, and deriving an amplitude of a sine wave component having the same frequency as the reference analog sine wave signal included in the input first measurement analog signal. A first amplitude deriving means for inputting a voltage induced by the third winding as a second measurement analog signal, and a sine having the same frequency as the reference analog sine wave signal included in the input second measurement analog signal. Nari A second amplitude deriving means for deriving the amplitude, e Bei a rotation angle calculating means for calculating a rotation angle of the rotor relative to the stator based on the amplitude calculated by said first and second amplitude deriving means And the first amplitude deriving means samples the input first measured analog signal at a predetermined rate.
1st measurement analyzer
First A / D conversion means for sequentially A / D converting a log signal
And a predetermined number obtained by A / D conversion by the first A / D conversion means.
Using the sampling data of
Sine wave function or cosine wave function with the same frequency as the sine wave signal
Apply the least squares method with the basis function as
Calculating a coefficient to obtain the first measured analog signal
And the same frequency as the reference analog sine wave signal.
With the first amplitude calculating means for calculating the amplitude of the sine wave component of the number
The second amplitude deriving means is configured to sample the input second measured analog signal at a predetermined rate.
2nd measurement analyzer
Second A / D conversion means for sequentially A / D converting log signals
And the location where the A / D conversion is performed by the second A / D conversion means.
Use constant sampling data and
Sine wave function or cosine wave of the same frequency as the analog sine wave signal
Applying the least squares method with the function as the basis function,
Calculating the coefficient of the number, the second measured analog
Same as the reference analog sine wave signal contained in the signal
Second amplitude calculating means for calculating the amplitude of the sine wave component of the frequency
And a rotation angle detecting device.