KR100827818B1 - Method for measuring the angles of rotatable bodies - Google Patents

Abstract

A method for measuring the angle of a rotary body is provided to implement simpler calculation by using fixed threshold value or variable threshold value according to a measurement error of relative rotation angle. A device for measuring the angle of a rotary body includes a main rotary body(11), a sub rotary body(13), first and second sensors(15,17), and an operator(19). The main rotary body rotates on its shaft powered by an external power supply. The sub rotary body rotates by interlocking with the main rotary body. The first sensor measures a relative rotation angle of the main rotary body. The second sensor measures a relative rotation angle of the sub rotary body. The operator calculates an absolute rotation angle of the main rotary body through operation based on data transmitted from the first and second sensors.

Description

Method for measuring the angles of rotatable bodies

1 is a view schematically showing an apparatus for measuring a shaft rotor configured to explain a method for measuring a rotation angle of the shaft rotor according to an embodiment of the present invention.

Figure 2 is a block diagram sequentially showing a method for measuring the rotation angle of the shaft rotating body according to an embodiment of the present invention.

3 is a diagram for explaining the basic principle of correcting the r value obtained in consideration of the relative rotation angle measurement error.

4 is a diagram illustrating a range of Φ _Ω / Ω values when the fractional value of r value obtained through calculation is 0.5 * r _e .

FIG. 5 is a diagram illustrating a range of Φ _Ω / Ω values when the decimal value of r value obtained through calculation is r _e + e / Ω.

FIG. 6 is a diagram illustrating a range of Φ _Ω / Ω values when the fractional value of r value obtained through calculation is r '.

7 is a schematic view of the shaft rotating body measuring apparatus shown in order to explain another example of a method for measuring the rotation angle of the shaft rotating body according to an embodiment of the present invention.

<Explanation of symbols for main parts of the drawings>

11: Main rotor 13: Auxiliary rotor

15: first sensor 17: second sensor

19: calculation unit 21: first auxiliary rotating body

23: second auxiliary rotating body

The present invention relates to a method for measuring the rotation angle of the shaft rotating body, and more particularly, the process of measuring the relative angle of rotation of a plurality of rotating body rotating in conjunction with a constant rotation ratio, thereby calculating the absolute angle of rotation of any rotating body It relates to a rotation angle measuring method of the shaft rotating body including a.

In recent years, automobiles are introduced with various high-tech devices for driving convenience and passenger safety. For example, the body position control device, called the electronic stability program (ESP), selectively brakes the wheels to prevent the vehicle from slipping during acceleration, braking or cornering of the vehicle, thereby stably positioning the vehicle and even making a driver's mistake. It is a system to calibrate.

In the vehicle body attitude control device, information such as the speed of the driving wheel, the braking pressure, the angle of rotation of the steering shaft, and the inclination of the vehicle body is detected through a sensor, and on the basis of this, the vehicle is driven by braking appropriate wheels among the wheels of the vehicle. Stabilize your car's posture to prevent accidents.

Prior arts related to a steering shaft rotation angle measuring device and a measuring method used for the vehicle body attitude control device are US Pat. Nos. 5,930,905 and 6,466,889. The prior art has a main rotor and two auxiliary rotors that rotate in conjunction with a constant rotation ratio in accordance with the rotation of the main rotor, and after measuring the relative rotation angle of these auxiliary rotors, the main rotor using this measurement value A method of calculating the absolute rotation angle of the system is presented.

In particular, in the US Patent No. 5,930,905, after measuring the relative rotation angle θ and ψ of the two auxiliary rotors, after calculating the value of k in [Equation 1] below, and then correcting it to an integer value, [Equation 2 ] Is used to calculate the absolute rotation angle φ of the main rotating body, and a process for additionally performing compensation for the negative value of φ calculated through the above process is disclosed.

[Equation 1]

k = [(m + 1) θ-mψ] / Ω

[Equation 2]

φ = [mψ + (m + 1) θ- (2m + 1) kΩ] / 2n

In the above [Formula 1] and [Formula 2], n, m and (m + 1) represent gear ratios of the three rotating bodies, and? Represents the period of the sensor. In particular, this method is applicable when the gear ratio of two auxiliary rotors is m: (m + 1).

In addition, in US Patent No. 6,466,889, when the main rotor rotates from the absolute rotation angle of 0 degrees to the maximum value, the value of φ c representing the difference between the absolute rotation angles of the two auxiliary rotors is continuously from 0 degrees to the sensor period Ω. The content which sets a rotation ratio so that it changes and calculates the value of (phi) from the relative rotation angle measurement value is disclosed.

When the absolute rotation angle has the maximum value, the auxiliary rotor rotates k ₁ times the sensor measurement period, and the maximum integer value smaller than the value obtained by multiplying the calculated value of Φ c by k ₁ / Ω is regarded as the rotation speed. To calculate the absolute rotation angle of this auxiliary rotor. Since the absolute rotation angle calculated in this way may include an error, it is necessary to finally check the possibility of occurrence of such an error and correct the calculated absolute rotation angle if necessary.

The present invention has a mechanism for measuring the relative rotation angle of a plurality of rotating bodies to rotate in conjunction with a constant rotation ratio and finds the absolute rotation angle of the desired rotating body through this, the calculation is simple and operates in the presence of a wide range of measurement error In addition to this, it is possible to optimally combine the maximum rotation range of the rotating body, the rotation ratio, the measurement cycle of the sensor, the allowable range of the relative rotation angle measurement error, and the error range of the calculated absolute rotation angle, as well as the relative rotation angle measurement. If the error is relatively small, the calculation amount can be minimized by using a fixed threshold value. If the relative rotation angle measurement error is relatively large, the allowable range of the relative rotation angle measurement error is increased as necessary by changing the threshold value. It allows you to make various choices, such as adjusting thresholds and related variables, to make the calculation simpler. Once in the whole providing a rotational angle measuring method of the object.

In order to achieve the above object, the present invention, when the main rotational body and the main rotational body in rotation with the main rotational body in a predetermined rotation ratio in conjunction with the main rotational body, the main rotational body rotates by Ω degree which is the measuring period of the sensor (Ω + d ), The first and second sensors, and the first and second sensors, which measure the rotation angles of the auxiliary rotating body, the main rotating body and the auxiliary rotating body, respectively, by which the measuring period is less than or equal to 360 degrees. A method for measuring the absolute rotation angle of the main rotor using a rotation angle measuring device having a calculation unit for calculating the internal rotation angle of the main rotor by receiving the measured value, the first and second sensors through the first and second sensors A relative rotation angle measuring step of measuring the relative rotation angles Φ _Ω and θ _Ω of the auxiliary rotating body, respectively; Calculate how many times the main rotor rotates the measuring cycle of the sensor based on the relative rotation angle (Φ _Ω ) of the main rotor and the relative rotation angle (θ _Ω ) of the auxiliary rotor obtained through the relative rotation angle measurement step. Calculating a main rotating body rotation cycle; Absolute rotation angle calculation step of obtaining the absolute rotation angle (Φ) of the main rotor by the relative rotation angle (Φ _Ω ) and the rotation period (i) of the main rotor obtained through the relative rotation angle measurement step and the main rotor rotation period calculation step Characterized in that comprises a.

Further, when the rotation ratio of the main rotating body and the auxiliary rotating body is 1 / n: 1 / m, the relative rotation angle of the main rotating body is Φ _Ω , the relative rotation angle of the auxiliary rotating body is θ _Ω , the d = (( nm) / m) * Ω, and the main circuit is full when the real number r indicating whether rotate several times the sensor measurement cycle (Ω) is, n> m, θ _Ω ≥Φ _Ω r = (θ _{Ω Ω} -Φ ) / d, θ _{Ω <Ω} Φ if = r (θ + _Ω Ω _Ω -Φ) / d, and, n <If Φ _{Ω Ω} ≥θ when _{m r = (Φ Ω -θ Ω} ) / (- d) If Φ _Ω <θ _Ω, r = (Φ _Ω + Ω-θ _Ω ) / (-d).

In addition, the step of performing after the main rotating body rotation period calculation step, if the error is included in the result of the main rotating body rotation period calculation further includes an error correction step of correcting the error to determine the correct number of rotation of the main rotating body It features.

In addition, the maximum value of the absolute error value of the first and second sensors is e ₁ , e ₂ , the arbitrary angle threshold value Φ _th is 0.5 * Ω, and the integer portion of the obtained r values is integer (r), decimal When the part is called fraction (r), the maximum error that r can have is r _e = (e ₁ + e ₂ ) / | d | I = integer (r) when (1) (r _e + e ₁ / Ω) ≤ fraction (r) <(1-r _e -e ₁ / Ω), and (2) 0 ≤ fraction (r) i = integer (r) if Φ _Ω <Φ _th if <(r _e + e ₁ / Ω), i = integer (r) -1 if Φ _Ω ≥ Φ _th and (3) (1-r _e- When e ₁ / Ω) ≤ fraction (r) <1, i = integer (r) + 1 when Φ _Ω <Φ _{th, and} i = integer (r) when Φ _Ω ≥ Φ _th .

Further, the angle threshold Φ _th is Φ _th = (fraction (r) +0.5) * Ω when (1) 0 ≤ fraction (r) <(r _e + e ₁ / Ω), and ( 2) When (1-r _e -e ₁ / Ω) ≤ fraction (r) <1, Φ _th = (fraction (r) -0.5) * Ω.

In addition, in setting the angular threshold value Φ _th , a fixed threshold method that always maintains the value of Φ _th at 0.5 * Ω, or (1) Φth when 0 ≤ fraction (r) <A Is (fraction (r) +0.5) * Ω, and (2) when B ≤ fraction (r) <1, Φ _th uses a variable threshold method that sets (fraction (r) -0.5) * Ω, In the actual calculation, when (Φ _th -δ) is used instead of the Φ _th value set in this way, and (Φ _th -δ) = Φ _{th '} , (1) when A ≤ fraction (r) <B, i Is i = integer (r) if Φ _Ω <Φ _{th '} for integer (r) (2) 0 ≤ fraction (r) <A, and i = integer (r) -1 for Φ _Ω ≥ Φ _th' (3) i = integer (r) +1 if Φ _Ω <Φ _{th 'for} B ≤ fraction (r) <1, i = integer (r) for Φ _Ω ≥ Φ _th' , and A value And B values are (r _e + e ₁ / Ω) ≤ A ≤ (Φ _th / Ω-δ / Ω-r _e -e ₁ / Ω), (Φ _th / Ω-δ / Ω + r _e + e ₁ / Ω) ≤ B ≤ (1-r _e -e ₁ / Ω) is characterized in that the value.

In addition, the range of the fraction (r) is divided into a predetermined number of sections, characterized in that the use of independent fixed or variable threshold value for each section.

In addition, when the measurement period of the first and second sensors are different, but the period (Ω ₁ ) of one sensor becomes an integer multiple of the period (Ω ₂ ) of the other sensor, the measurement from the sensor having a period of Ω ₁ The remaining value divided by Ω ₂ is used as a measurement of a sensor having a period of Ω ₁ .

Hereinafter, one embodiment according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a view schematically showing an apparatus for measuring a shaft rotor configured to explain a method for measuring a rotation angle of the shaft rotor according to an embodiment of the present invention.

Referring to the drawings, the axial rotating body measuring apparatus includes a main rotating body 11 which is axially rotated by external power, an auxiliary rotating body 13 which rotates in conjunction with the rotation of the main rotating body 11, and the main rotating body ( 11, from the first sensor 15 for measuring the relative rotation angle, the second sensor 17 for measuring the relative rotation angle of the auxiliary rotating body 15, and the first and second sensors 15, 17. It has a configuration including a calculation unit 19 for receiving the detected content to find out the absolute rotation angle of the main rotating body 11 through the internal calculation.

For reference, the absolute rotation angle means [(total rotation speed of the rotating body * 360 degrees) + the angle after the last wheel more rotated]. For example, if the rotor rotates one and half turns, the absolute rotation angle is 540 degrees. Relative rotation angle is also measured from the sensor, which is the remainder of the sensor's measurement period divided by the angle turned further after the last wheel. For example, if the absolute rotation angle is 400 degrees, the relative rotation angle is 40 degrees in both cases where the measuring period of the sensor is 360 degrees or 180 degrees. However, if the absolute rotation angle is 600 degrees, the relative rotation angle is 240 degrees when the measuring period of the sensor is 360 degrees, the relative rotation angle is 60 degrees when the measuring period of the sensor is 180 degrees.

In addition, although the main rotating body 11 and the auxiliary rotating body 13 takes the form of a cylinder or a disk in FIG. 1, a gear may be applied in some cases.

For the following description, the maximum measurable angle (hereinafter, the measuring period) of the first and second sensors 15 and 17 is set to Ω, and the rotation ratio of the main rotating body 11 and the auxiliary rotating body 13 is 1. Let it be / n: 1 / m. When the main rotating body 11 and the auxiliary rotating body 13 is a gear, the number ratio of the main rotating body 11 and the auxiliary rotating body 13 will be n: m.

In addition, the absolute rotation angle and the relative rotation angle of the main rotating body 11 are represented by Φ and Φ _Ω , respectively, and the relative rotation angle of the auxiliary rotating body is represented by θ and θ _Ω , respectively. The final goal of this embodiment is to calculate Φ by measuring Φ _Ω and θ _Ω . In particular, the rotation ratio of the main rotating body 11 and the auxiliary rotating body 13 is designed so that the difference between Φ and θ is not larger than Ω even if the main rotating body 11 is at any position within the operating range.

After measuring Φ _Ω and θ _Ω from the first and second sensors 15 and 17, the process of calculating Φ based on this is composed of several steps shown in FIG. 2. That is, the relative rotation angles of the main rotating body 11 and the auxiliary rotating body 13 are measured using the first and second sensors, and a real value r indicating how many times the measuring cycle of the sensor rotates is measured. Computing the error included in the result of the main rotating body rotation period calculation to determine the correct number of revolutions (i) of the main rotating body, and finally to determine the absolute rotation angle (Φ) of the main rotating body.

With the above matters in mind, a method of measuring the rotation angle of the shaft rotating body according to the present embodiment will be described in earnest.

First, when the relative rotation angle of the main rotating body 11 is Φ _{Ω as} a result of measuring through the first sensor 15, the main rotating body 11 rotates i times the measuring period (Ω) of the sensor by Φ _Ω It can be seen that it is rotated more. (Of course, one rotation may not have been performed. In this case, i = 0) Therefore, the absolute rotation angle Φ of the main rotor 11 can be expressed by the following general formula.

Φ = i * Ω + Φ _Ω ... [Equation (1)]

If the integer value i can be calculated in [Equation (1)], the value of Φ is calculated immediately. In this embodiment, the calculation of the integer value i is performed in two steps.

In the first step, a real value r indicating how many times the main rotating body 11 rotates the sensor measurement period Ω is calculated using the Φ _Ω value and the θ _Ω value.

Assuming that n is larger than m, when the main rotor 11 rotates by one times the sensor measurement period (Ω), the auxiliary rotor rotates one times the measurement period (Ω) and then Rotate the same angle d further.

That is, d = ((n-m) / m) * Ω ... [Equation (2)]

If the value of d is obtained through the above [Equation (2)], the real value r indicating how many times the main rotating body 11 rotates the sensor measurement period (Ω) can be obtained using the following [Equation (3)]. have.

r = (θ-Φ) / d ... [Equation (3)]

In the formula (3), θ and Φ are values that have not yet been calculated. However, if one condition is met, the difference between the two values can be calculated without knowing each of these values. Specifically, the following relationship is obtained from the condition that the difference between θ and Φ is always less than Ω and [Equation (1)]:

Φ = i * Ω + Φ _Ω ≤ θ <(i + 1) * Ω + Φ _Ω = Φ + Ω ... [Equation (4)]

As expressed by Equation (4), when the main rotor 11 rotates i times the sensor measurement period Ω and rotates by Φ _Ω , the auxiliary rotor 13 measures the sensor. You may not have been able to rotate (i + 1) times the period (Ω), or you may have entered the (i + 2) th revolution by rotating (i + 1) times the sensor measurement period (Ω). have.

If the auxiliary rotor 13 does not complete the (i + 1) th rotation, θ _Ω has a larger value than Φ _Ω , and the auxiliary rotor 13 enters the (i + 2) th rotation. If you stand, θ _Ω will be smaller than Φ _Ω . Therefore, we can rewrite Equation (3) as follows using relative rotation angle.

If θ _Ω ≥ Φ _Ω , r = (θ _Ω -Φ _Ω ) / d

If θ _Ω <Φ _Ω , r = (θ _Ω + Ω-Φ _Ω ) / d ... [Equation (5)]

In the above [Equation (5)], θ _Ω and Φ _Ω have the same value, which corresponds to the case where the absolute rotation angle of the main rotor 11 is 0 degrees.

In setting the values of n and m that determine the rotation ratio of the main rotor and the auxiliary rotor, the value of (θ-Φ) should be smaller than Ω when the absolute rotation angle of the main rotor is the maximum value, but designed to be as close as possible to Ω. It is advantageous in terms of increasing the allowable range of the relative rotation angle measurement error, which will be described later. However, if the value of (θ-Φ) is set so close to Ω, the value of Φ is the maximum value during the calculation of [Equation (5)] due to the error in the relative rotation angle measurement. The case of 0 degrees may not be distinguishable. Therefore, even if the absolute rotation angle of the main rotor reaches the maximum value, the values of n and m are determined by allowing design to allow the difference between θ and Φ to be somewhat smaller than Ω.

When the value of r is obtained through the above process, the value of Φ may be calculated by substituting the value of r into Equation (6) below.

Φ = r * Ω ... [Equation (6)]

However, the calculation method according to [Equation (6)] can be applied only when the relative rotation angle (θ _Ω and Φ _Ω ) measurement errors of the sensor are very small. Usually, the relative rotation angle measurement values Φ _Ω and θ _Ω used to calculate the value of r may include errors inherent to the sensors 15 and 17 or electronic, mechanical and optical errors. Therefore, the absolute rotation angle Φ of the main rotating body 11 obtained through the above [Equation (6)] is very likely to contain a considerable degree of error.

If the first and second sensors 15 and 17 indicate the maximum value of the absolute value of the error of measuring the relative rotation angles of the main rotating body 11 and the auxiliary rotating body 13 as e, [Equation 5] The maximum value of the absolute error value (r _e ) that can be included in the value of r calculated in [] is calculated as follows.

r _e = 2 * e / d ... [Equation (7)]

For example, if the value of e is 1 degree, the value of d is 30 degrees, and the measuring period of the sensor is 180 degrees, the value of r _e is 2/30, and the value of Φ calculated by [Equation (6)] is calculated. The maximum error is 12 degrees. If the above error is allowed, the value Φ calculated by substituting the r value in [Equation (6)] can be used. If the error is not allowed, the error should be corrected.

That is, the error correction is performed on the value of r using information on the relative rotation angle measurement error (specifically, the maximum value of the measurement error, that is, the value of e), and through this, how many times the main rotor rotated over the sensor period. Compute the integer value i with no error indicated. When the value of i is correctly determined, the value of the absolute rotation angle Φ of the main rotor 11 can be calculated by [Equation (1)].

In the meantime, the integer part of the calculated r value is represented by integer (r) and the fractional part by fraction (r). Where fraction (r) is a positive real number greater than or equal to zero and less than one. For example, if r is 1.2257, interger (r) is 1 and fraction (r) is 0.2257.

The calculation of the integer value i by correcting the r value obtained as described above is performed by dividing into three cases as shown in FIG. 3 according to the value of fraction (r).

Referring to FIG. 3, it can be seen that the fraction (r) value increases in the counterclockwise direction. The fraction (r) value increases in a counterclockwise direction as the main rotor 11 rotates in a direction in which the absolute rotation angle increases, and then decreases to a value of 0 when the value is equal to 1, and then increases again. Here, the value of the fraction (r) starts from 0, increases, and then returns to 0 means that the main rotor 11 is rotated by the sensor measuring period (Ω). The fraction (r) value shown in FIG. 3 can be viewed as a relative value obtained by dividing the rotation angle of the main rotor 11 by Ω.

In any case, of the above three cases, the first case is a case where the fraction (r) at the obtained r value is greater than or equal to (r _e + e / Ω) and less than (1-r _e -e / Ω). That is, in the case of (r _e + e / Ω) ≦ fraction (r) <(1-r _e -e / Ω), it is the region I of FIG. 3.

Since the calculated maximum error of r is r _e , when fraction (r) is in region I, the actual value of fraction (r) is greater than or equal to e / Ω and less than (1-e / Ω). Have Here, the actual value is a value when it is assumed that there is no relative rotation angle measurement error. In this case, even though the measured value Φ _Ω includes an error of up to e, the effect of this error on the value of r is smaller than e / Ω, so that the value of i is calculated as it is in the formula (1). When set to (r), accurate calculations are made. In other words,

i = integer (r) ... [expression (8)]

In the second case, the fraction (r) value is greater than or equal to 0 and smaller than (r _e + e / Ω), which is the region II of FIG. 3. That is, when 0≤fraction (r) <(r _e + e / Ω). When the value of fraction (r) is in zone II, the value of i may be equal to integer (r) or i may have a value of (integer (r) -1). In other words, the value r is calculated as the main rotor 11 has already rotated the integer (r) times of the sensor measurement period (Ω), but in reality, the main rotor 11 is the integer (r) times of the sensor period. May not have rotated). This phenomenon is caused by the error included in the relative rotation angle measurement.

Zone III will be described later.

FIG. 4 is a diagram illustrating a possible range of Φ _Ω / Ω values when the fraction r in the calculated r value is 0.5 * r _e . As described above, since the magnitude of the error included in the calculated r value is at most r _e , the actual value of r is between (integer (r) -0.5 * r _e ) and (integer (r) + 1.5 * r _e ). Exists in The value of i is either integer (r) or (integer (r) -1).

As described above, since i may be interger (r) or (interger (r) -1), one of the two values should be determined. This determination can be made by using the relative rotation angle Φ _Ω of the main rotor 11.

That is, considering the relative rotation angle measurement error as shown in Figure 4, Φ _Ω is less than (1.5 * r _e * Ω + e) or ((1-0.5 * r _e ) * Ω-e) Has a large value. By considering the existence range of such Φ _Ω value, has the slightly larger measure Φ _Ω is greater than zero which determines hayeotdago already rotating the integer (r) times (倍) or more of the entire main sensor measurement period (Ω) state [ Equation (1)] means that the calculation can be performed. On the other hand, if Φ _Ω has a value close to Ω, this means that the main rotor has not yet rotated by an integer (r) times the sensor measurement period (Ω) and can perform calculation of [Equation (1)]. .

On the other hand, that the value of Φ _Ω as described above, a little greater than or equal to Φ _Ω is close proximity to Ω in accordance with the result of comparison in comparison because ambiguity, a certain threshold value or appointed the reference value Φ _Ω value to a threshold value Determine the value of i. In this description, the above threshold is expressed as φ _th .

The threshold value Φ _th is preferably 0.5 * Ω in consideration of the symmetry of the sensor error range (regions II and III are symmetric in Fig. 3) and the allowable range of the relative rotation angle measurement error to be described later. You can also use other values if you need to.

In any case, by introducing the threshold value φ _th , the exact value of the constant i can be determined by the following criteria even in the presence of the relative rotation angle measurement error.

I = integer (r) if Φ _Ω <Φ _th

If Φ _Ω ≥Φ _th, i = integer (r) -1 ... [Equation (9)]

3 again, if the fraction (r) value is greater than or equal to (1-r _e -e / Ω) and less than 1, i.e., (1-r _e -e / Ω) ≤fraction (r) <1 In this case (zone III), i may have a value of integer (r) or (integer (r) +1). In other words, the r value was calculated as that the main rotor 11 failed to rotate (integer (r) +1) times the sensor measurement period Ω, but in reality, the main rotor had a sensor measurement period Ω. The (integer (r) +1) times (倍) may have already been rotated. This phenomenon occurs because of the errors in the relative rotation angle measurements.

In this case, through a similar development process as in section II above, the value of i (even in the presence of relative rotation angle measurement errors) can be accurately determined according to the following criteria:

If Φ _Ω <Φ _th , i = integer (r) +1

If Φ _Ω ≥Φ _th , i = integer (r) ... [Equation (10)]

As in the case of the above zone II, the value of Φ _th in the zone III is preferably 0.5 * Ω.

Subsequently, a process of calculating the value of the absolute rotation angle Φ of the main rotor using the above calculation result is performed. Since the correct i value was found through the above process, the main rotor rotated i times (倍) of the sensor measurement period (Ω) and rotated again by Φ _Ω , so the value of Φ is expressed by [Equation (1)]. Can be calculated by Since only Φ _Ω has an error component at the right side of [Equation (1)], the error included in the calculated absolute rotation angle Φ of the main rotor is equal to the error included in the Φ _Ω value. After the calculation through [Equation (1)], measure the relative rotation angle of the rotating bodies again and repeat the process of calculating r value.

The method proposed in this embodiment makes it possible to accurately calculate the value of the constant i, which indicates how many times the main rotor 11 has rotated more than the sensor measurement period Ω even if there is an error in the measurement of the relative rotation angle. If so, it is important to know how much the relative rotation angle measurement error can tolerate.

The maximum value of the absolute value of the relative rotation angle measurement error that the proposed method can tolerate is zone II of FIG. 3 when the fraction (r) value in the calculated r value is smaller than (r _e + e / Ω). It can be calculated using the example of. It is assumed that the value of fraction (r) is slightly smaller than (r _e + e / Ω), but when calculating the tolerance, the value of fraction (r) is equal to (r _e + e / Ω). Not only can the result be obtained, but the calculation is convenient.

However, since the calculated value of r may be different by r _{e due} to the relative rotation angle measurement error, as shown in FIG. 5, the actual value of fraction (r) is from e / Ω to (2 * r _e + e / Ω). Exists between). This is the range of the actual value of the relative rotation angle of the main rotor 11. However, we cannot accurately measure the relative rotation angle of the main rotor 11 due to the measurement error, and the value we measure is Φ _Ω added to the measurement error by a maximum of e.

Thus, as shown in FIG. 5, Φ _Ω is greater than or equal to 0 and has a value less than (Ω * 2 * r _e + 2 * e). Since fraction (r) is a relative value between 0 and 1 indicating how many times the sensor measurement period (Ω), FIG. 5 shows a range of values of Φ _Ω divided by Ω.

Therefore, in the case of FIG. 5, if the value of the threshold value φ _th is set to be larger than (Ω * 2 * r _e + 2 * e) and smaller than Ω, the value of i is accurately determined using the above-described formula (9). Can be calculated If Φ _th is fixed to have a constant value at all times, the calculation becomes simple, and this method is called a fixed threshold method. In order to maximize the allowable range of the relative rotation angle measurement error in the fixed threshold method, it is reasonable to set Φ _th to 0.5 * Ω in consideration of the symmetry of the zone II and zone III. At this time, the maximum value of the allowable relative rotation angle measurement error is calculated by the above-described formula (7) and the following formula (11).

Ω * 2 * r _e + 2 * e = 0.5 * Ω ... [Equation (11)]

As can be seen from Fig. 5, if the relative rotation angle measurement error is within the allowable range indicated by [Equation (11)], φ _th is set to 0.5 * Ω no matter what the calculated fraction (r) has in the II zone. Equation (9) can be used to calculate the value of i accurately.

In addition, as shown in FIG. 3, since the zone III is symmetrical with the zone II, the maximum allowable range of the threshold value and the relative rotation angle measurement error in the zone III can be calculated by the same method, and the same as in the zone II. The calculation result is obtained.

When the value of e in the actual application environment using the fixed threshold method is smaller than the maximum allowable range of the error given in [Equation (11)], in the case of zone II of FIG. 3, (2 * r _e + 2 * e) / Ω) is less than 0.5, so the actual angle threshold can be lowered from 0.5 * Ω to (2 * r _e * Ω + 2 * e). And considering the symmetrical zone III, the threshold is randomized between (2 * r _e * Ω + 2 * e) and (Ω-2 * r _e * Ω-2 * e) instead of 0.5 * Ω. Can be used as a value.

If the angle threshold is determined to be (0.5 * Ω-δ) instead of 0.5 * Ω, the boundary between the I and II zones is calculated from (r _e + e / Ω) to (0.5-δ / Ω- in the actual calculation. r _e -e / Ω). Because when the value of fraction (r) is between (r _e + e / Ω) and (0.5-δ / Ω-r _e -e / Ω), the relative rotation angle measurement Φ _Ω is always greater than or equal to 0 degrees This is because it becomes smaller than the threshold value and the same result is obtained even when using any of [Expression (8)] and [Expression (9)]. Similarly, the boundary between zones I and III can be used by selecting any value between (0.5-δ / Ω + r _e + e / Ω) to (1-r _e -e / Ω). have.

If there is a certain degree of freedom in setting the angle threshold and the boundary between the zones, the angle threshold and the boundary line between the zones can be selected so that a comparison operation for obtaining an i value can be implemented in a simpler form. As the actual threshold used in the fixed threshold method drops away from 0.5 * Ω, and the larger the area of the II and III zones than the minimum required, the maximum allowable range of the resulting angle measurement error is reduced. The range may be calculated by the same method as described with reference to FIG. 5.

On the other hand, if the relative rotation angle measurement error satisfies the allowable range calculated using [Equation (11)], it is most efficient to use the fixed threshold method. However, depending on the application environment, the relative rotation angle measurement error may exceed the allowable range indicated by [Equation (11)]. In such a case, the maximum allowable range of the relative rotation angle measurement error can be extended by changing the value of φ _th according to the fraction (r) value in the calculated r value. This method is called a variable threshold method.

As shown in FIG. 6, the error tolerance in the variable threshold method can be calculated using an example in which the fraction (r) in the calculated r value is r '. Since r 'is smaller than re in FIG. 6, this case corresponds to region II of FIG. 3.

If the relative rotation angle measurement error is large, the value of r _e is also increased by [Equation (7)], and as shown in FIG. 6, the range of the value of _Ω is greatly expanded. At this time, if the value of Φ _th is 0.5 * Ω, it is impossible to calculate the exact value of i through [Equation (9)], but this problem can be solved by changing the value of Φ _{th according} to the calculated fraction (r) value. .

For example, in the case of FIG. 6, the value of Φ _th should be set larger than 0.5 * Ω. At this time, the value of Φ _th can be set in various forms, but in order to maximize the allowable range of the relative rotation angle measurement error, it is reasonable to set the value of Φ _th as follows.

Φ _th = (fraction (r) +0.5) * Ω ... [Equation (12)]

At this time, the maximum value of the allowable relative rotation angle measurement error is calculated by [Equation (7)] and [Equation (13)] below.

Ω * 2 * r _e + 2 * e = Ω ... [Equation (13)]

This variable threshold method is always applicable when the fraction (r) in the calculated r value exists in the II region. When the variable threshold method is used, the process of calculating φ _th is added, compared to when using the fixed threshold method described above, while the allowable range of the relative rotation angle measurement error is doubled.

The method of increasing the allowable range of the relative rotation angle measurement error using the variable threshold method is equally applicable to the III zone. For Zone III, Φ _th can be set as follows through the same process as for Zone II.

Φ _th = (fraction (r) -0.5) * Ω ... [Equation (14)]

Even in the variable threshold method, when the e value in the actual application environment is smaller than the maximum allowable range of the error given in [Equation (13)], the degree of freedom in setting the angle threshold and the boundary between the regions can be given. These values can be selected to make the comparison operation required for calculating the values simpler.

If the variable threshold is determined to use (Φ _th -δ) instead of the Φ _th value calculated as [Eq. (12)] or [Eq. (14)], then an analysis similar to that performed in the fixed threshold method is performed. The boundary between zones I and II can be set to any value between (r _e + e / Ω) and (Φ _th / Ω-δ / Ω-r _e -e / Ω). In addition, the boundary line between Zone I and Zone III can be used by selecting any value between (Φ _th / Ω-δ / Ω + r _e + e / Ω) and (1-r _e -e / Ω).

Depending on the application environment, a combination of a fixed threshold method and a variable threshold method may be used. For example, the range of the fraction (r) value may be divided into a predetermined number of sections, and an independent fixed or variable threshold method may be used for each section.

On the other hand, the above calculation method can be applied even when n has a value smaller than m. In this case, the roles of the main rotor and the auxiliary rotor may be changed. However, in this case, the error is amplified by multiplying m / n in the process of finally changing the absolute rotation angle of the auxiliary rotor to the absolute rotation angle of the main rotor.

In order to prevent the amplification of such an error, when the value of n is smaller than the value of m, when the main rotor rotates by Ω, think about how much less than the Ω rotates, it is possible to perform the calculation using this. Specifically, the above-mentioned [formula (3)], [formula (5)], and [formula (7)] are respectively changed as follows, and the calculation becomes possible.

r = (Φ-θ) / (-d) ... [Equation (15)]

If Φ _Ω ≥ θ _Ω , r = (Φ _Ω -θ _Ω ) / (-d)

If Φ _Ω <θ _Ω , r = (Φ _Ω + Ω-θ _Ω ) / (-d) ... [Equation (16)]

r _e = 2 * e / (-d) ... [Equation (17)]

After all, in the present invention, the name of the main rotating body or the auxiliary rotating body is attached for convenience of description, and the difference between the absolute rotation angle of the two rotating bodies is smaller than the sensor period in a general situation in which there are two rotating bodies rotating in conjunction with a constant rotation ratio. If the condition is satisfied, in the present invention, any one of the two rotating bodies can be regarded as the main rotating body to perform the absolute rotation angle calculation.

In the present invention, this property can be used to increase the accuracy of the final calculation result. First, the absolute rotation angle Φ of the main rotating body 11 is calculated through the above calculation with respect to the main rotating body 11. Then, after performing the above-described calculation with respect to the auxiliary rotating body 13, the absolute rotating angle Φ 'of the main rotating body is calculated again using the calculated absolute rotating angle θ and the rotating ratio of the auxiliary rotating body.

The Φ includes an error according to the relative rotation angle measurement of the main rotating body 11, and Φ 'includes a relative rotation angle measurement error of the auxiliary rotating body 13 converted into a rotation ratio. Then, after calculating the variance of these two error components, calculate the average of them by appropriately weighting Φ and Φ 'based on them, and using the obtained average as the absolute rotation angle of the main rotor, the influence of the measurement error Can reduce the number and increase the accuracy of the operation.

Application systems to which the present invention is applied can diagnose failures of the angle measuring device and associated data processing device in various forms. For example, it is possible to determine whether or not the related device is broken by using whether or not the measured value from the angle measuring device exceeds the range of the normal angle change rate. Or failure of the angle measuring device and associated processing device by checking whether the real value r and the integer value i or the final absolute rotation angle, respectively, indicating how many times the rotating body has rotated the sensor measurement cycle exceed the normal range of change. Can be determined. If a fault is found in one angle sensor and its associated processing unit, the power supply may be turned on using another sensor and its Normal operation can continue as long as it is supplied. Thus, estimating the value of current i using past i values without having to recalculate the value of i each time has the effect of reducing the amount of computation and also for normal operation under transient failures such as unexpected loud noises. It can be used as.

On the other hand, the present invention can be applied in various forms depending on the application environment. First, depending on the application environment, it may be difficult or uneconomical to directly measure the relative rotation angle of the main rotor. In this case, additional auxiliary rotors can be used. The additional auxiliary rotor may be rotated in engagement with the main rotating body or may be rotated in engagement with the existing auxiliary rotating body.

7 is an application example in which the main rotor and the two auxiliary rotors rotate in conjunction with a rotation ratio of 1 / p: 1 / n: 1 / m, respectively. In FIG. 7, the first auxiliary rotating body 21 and the second auxiliary rotating body 23 rotate in conjunction with the main rotating body 11. The absolute rotation angle of the main rotating body 11 is ψ, and the absolute rotation angle of the first auxiliary rotating body 21 is φ. The tooth ratio is p: n: m when the gear ratio of the main rotor 11, the first auxiliary rotor 21, and the second auxiliary rotor 23 are implemented in gears.

In general, when three or more rotors are interlocked, after measuring the relative rotation angle of the two rotors of these rotors, after calculating the absolute angle of rotation of one of the two rotors through the above calculation method, The rotation ratio can be used to calculate the absolute rotation angle of the desired rotating body.

In the case of FIG. 7, the relative rotation angles of the first auxiliary rotating body 21 and the second auxiliary rotating body 23 are measured through the first and second sensors 15 and 17. 1 The absolute rotation angle of the auxiliary rotating body 21 can be calculated.

When the p and n are designed to have the same value, the absolute rotation angle of the main rotating body 11 is equal to the absolute rotation angle of the first auxiliary rotating body 21. However, p and n do not necessarily have the same value, and it may be advantageous to set different values of p and n depending on the application environment.

If p and n are set to have different values, the absolute rotation angle Ψ of the main rotor can be obtained by the following formula (18).

Ψ = Φ * (n / p) ... [Equation (18)]

If p is designed to have a value larger than n, the size of the first auxiliary rotor 21 is reduced and the error included in the value of Ψ is reduced to n / p compared to the error included in the value of Φ. However, the range of the absolute rotation angle of the first subsidiary rotating body 21 is increased compared to the main rotating body 11, and n and m should be set so that the magnitude of the value of d represented by [Equation (2)] becomes smaller. Therefore, the allowable range of the relative rotation angle measurement error expressed by [Equation (11)] and [Equation (13)] is reduced. Conversely, if p is designed to have a value less than n, the opposite effect is obtained compared to when p is greater than n.

Meanwhile, in the above description, it is assumed that the measurement periods (Ω) of the two sensors 15 and 17 used for the relative rotation angle measurement are the same. However, depending on the application environment, angle sensors with different measuring cycles may be used for each rotor. Here, when the two displays the sensor measurement cycle of each Ω ₁ and Ω _2, Ω ₁ to remainders obtained by dividing the measured value to Ω ₂ from the sensor has the value of an integral multiple of Ω _2, having a period of Ω ₁ by considered to be the measured value of the sensor with a period of ₁ Ω, it is possible to perform the operation as if both the period of the two sensors ₂ Ω.

When using different types of angle sensors, the error characteristics included in the relative rotation angle measurements from the two sensors may be different. Sometimes, the same kind of sensor can cause a similar situation due to a number of factors. If the maximum value of the absolute value of the error included in the measured relative rotation angles from the two sensors is expressed as e ₁ and e ₂ , respectively, in this case, replace [Equation (7)] and [Equation (17)] as follows. Just do it.

r _e = (e ₁ + e ₂ ) / | d | ... [Equation (19)]

In this case, e contained in the second term on the left side in [Equation (11)] and [Equation (13)] is replaced with one of e ₁ and e ₂ . Specifically, among the e ₁ and e _2, one representing the error characteristic with respect to the measured value of Φ _Ω used in the calculation of the absolute rotation angle of the main rotating body 11 is selected and used instead of e.

As mentioned above, although this invention was demonstrated in detail through the specific Example, this invention is not limited to the said Example, A various deformation | transformation is possible for a person with ordinary knowledge within the scope of the technical idea of this invention.

In particular, depending on the application environment, a larger number of auxiliary rotors and angle sensors than those shown in FIG. 1 or FIG. 7 may be used. For example, a plurality of auxiliary rotors may further be used as a means for achieving a desired rotation ratio. Alternatively, in order to increase the reliability, it is possible to have several angle sensors on one rotor or to use an additional auxiliary rotor and a pair of angle sensors. In addition, in order to reduce the error due to the measurement of the relative rotation angle, it is possible to have several angle sensors on one rotor or to use an additional auxiliary rotor and a pair of angle sensors. For example, it is possible to reduce the variance of the relative rotation angle measurement error by placing several angle sensors on one rotating body and using the weighted average value of the measurements from these sensors as the relative rotation angle measurement of the rotating body.

The present invention made as described above, has a mechanism for measuring the relative rotation angle of a plurality of rotating body rotating in conjunction with a constant rotation ratio and find out the absolute rotation angle of the desired rotating body through this, the calculation is simple and a wide range of measurement error Not only is it possible to operate in the presence of, but also the optimal combination of the maximum rotation range of the rotor, the rotation ratio, the measuring cycle of the sensor, the allowable range of the relative rotation angle measurement error and the calculated absolute rotation angle error range, If the relative rotation angle measurement error is relatively small, the calculation amount can be minimized by using a fixed threshold value. If the relative rotation angle measurement error is relatively large, the allowable range of the relative rotation angle measurement error can be set as necessary by changing the threshold value. Increasing thresholds and adjusting thresholds and related variables to make calculations simpler. It enables the selection.

Claims (9)

A main rotating body capable of shaft rotation, and an auxiliary rotating body which rotates (Ω + d) when the main rotating body rotates at a predetermined rotation ratio in conjunction with the main rotating body at a predetermined rotation ratio, and the main rotating body rotates by Ω degrees, which is a measurement period of the sensor; By measuring the rotation angle of the main rotor and the auxiliary rotor, respectively, the measurement period (Ω) is less than or equal to 360 degrees, the first and second sensors and the value measured by the first and second sensors are received internally calculated As a method of measuring the absolute rotation angle of the main rotor using a rotation angle measuring device having a calculation unit for finding the rotation angle of the main rotor, A relative rotation angle measuring step of measuring relative rotation angles (Φ _Ω and θ _Ω ) of the main and auxiliary rotors through the first and second sensors, respectively; Calculate how many times the main rotor rotates the measuring cycle of the sensor based on the relative rotation angle (Φ _Ω ) of the main rotor and the relative rotation angle (θ _Ω ) of the auxiliary rotor obtained through the relative rotation angle measurement step. Calculating a main rotating body rotation cycle; And an absolute rotation angle calculation step of obtaining an absolute rotation angle of the main rotation body through the relative rotation angle (Φ _Ω ) and the rotation period (i) of the main rotation body obtained through the relative rotation angle measurement step and the main rotation body rotation period calculation step. , When the rotation ratio of the main rotor and the auxiliary rotor is 1 / n: 1 / m, the relative rotation angle of the main rotor is Φ _Ω , the relative rotation angle of the auxiliary rotor is θ _Ω , D = ((n-m) / m) * Ω, The real value r indicating how many times the main rotor rotated the sensor measurement period (Ω) is when n> m, if θ _Ω ≥Φ _Ω r = (θ _Ω -Φ _Ω ) / d, r = (θ _Ω + Ω-Φ _Ω ) / d if θ _Ω <Φ _Ω , when n <m, If Φ _Ω ≥θ _Ω, r = (Φ _Ω -θ _Ω ) / (-d) If Φ _Ω <θ _Ω, r = (Φ _Ω + Ω-θ _Ω ) / (-d), The step of performing after the main rotating body rotation period calculation step, if the error is included in the result of the main rotating body rotation period correction error correction step for correcting the error to determine the exact number of revolutions (i) of the main rotating body further comprises: Method for measuring the rotation angle of the shaft rotor, characterized in that. delete delete The method of claim 1, The maximum value of the absolute error value of the measured values from the first and second sensors is e ₁ , e ₂ , and the arbitrary angle threshold Φ _th is 0.5 * Ω, and the integer portion of the obtained r values is integer (r), When the fractional part is called fraction (r), The maximum error that r can have is r _e = (e ₁ + e ₂ ) / | d | Is, (1) i = integer (r) when (r _e + e ₁ / Ω) ≤ fraction (r) <(1-r _e -e ₁ / Ω), (2) When 0 ≤ fraction (r) <(r _e + e ₁ / Ω) I = integer (r) if Φ _Ω <Φ _th , i = integer (r) -1 if Φ _Ω ≥ Φ _th , (3) (1-r _e -e ₁ / Ω) ≤ fraction (r) <1 I = integer (r) + 1 when Φ _Ω <Φ _th, i = integer (r) when Φ _Ω ≥ Φ _th . The method of claim 4, wherein The angle threshold Φ _th is, (1) Φ _th = (fraction (r) +0.5) * Ω when 0 ≤ fraction (r) <(r _e + e ₁ / Ω), (2) When (1-r _e -e ₁ / Ω) ≤ fraction (r) <1, Φ _th = (fraction (r) -0.5) * Ω. The method of claim 4, wherein When the angle threshold is used as (Φ _th −δ) instead of Φ _th value, and (Φ _th −δ) = Φ _{th '} , (1) If A ≤ fraction (r) <B, i is integer (r) (2) when 0 ≤ fraction (r) <A, i = integer (r) if Φ _Ω <Φ _{th '} and i = integer (r) -1 if Φ _Ω ≥ Φ _th' , (3) When B ≤ fraction (r) <1, i = integer (r) +1 if Φ _Ω <Φ _{th ',} i = integer (r) if Φ _Ω ≥ Φ _th' , The A value and the B value, (r _e + e ₁ / Ω) ≤ A ≤ (Φ _th / Ω-δ / Ω-r _e -e ₁ / Ω), (Φ _th / Ω-δ / Ω + r _e + e ₁ / Ω) ≤ B ≤ (1-r _e -e ₁ / Ω) is a value that satisfies the rotation angle measuring method of the rotating body. The method of claim 4, wherein The angle threshold Φ _th is, (1) When 0 ≤ fraction (r) <A, Φ _th is (fraction (r) +0.5) * Ω, (2) When B ≤ fraction (r) <1, Φ _th is (fraction (r) -0.5) * Ω, In actual calculation, when the angle threshold is used as (Φ _th −δ) instead of Φ _th value, and (Φ _th −δ) = Φ _{th '} , (3) If A ≤ fraction (r) <B, i is integer (r) (4) When 0 ≤ fraction (r) <A, i = integer (r) if Φ _Ω <Φ _{th '} and i = integer (r) -1 if Φ _Ω ≥ Φ _th' , (5) When B ≤ fraction (r) <1, i = integer (r) +1 if Φ _Ω <Φ _{th 'and} i = integer (r) if Φ _Ω ≥ Φ _th' , The A value and the B value, (r _e + e ₁ / Ω) ≤ A ≤ (Φ _th / Ω-δ / Ω-r _e -e ₁ / Ω), (Φ _th / Ω-δ / Ω + r _e + e ₁ / Ω) ≤ B ≤ (1-r _e -e ₁ / Ω) is a value that satisfies the rotation angle measuring method of the rotating body. The method according to any one of claims 4 to 7, And dividing the range of the fraction (r) into a predetermined number of sections and using independent fixed or variable threshold values for each section. The method of claim 1, When the measurement periods of the first and second sensors are different but the period (Ω ₁ ) of one sensor becomes an integer multiple of the period (Ω ₂ ) of the other sensor, the measured value from the sensor having a period of Ω ₁ is obtained. A method of measuring the rotation angle of a shaft rotor, characterized in that the remaining value divided by Ω ₂ is used as a measurement value of a sensor having a period of Ω ₁ .

Images