阅读说明：本技术 用于转速测量的梯度确定 (Gradient determination for rotational speed measurement ) 是由 安德烈亚斯·文德策尔 拉尔夫·德雷布霍尔茨 于 2018-11-07 设计创作，主要内容包括：本发明涉及具有能转动的机构、传感器和评估单元的装置,其中,机构具有数量为a的标记,其中,当机构转动时,标记循环地经过由传感器检测的区域；其中,传感器构造成用于将信号传输至评估单元；并且其中,评估单元构造成用于给每个信号配属传输信号的时间点t<Sub>i</Sub>,其中i∈{0,1,…}并且对于所有i∈{0,1,…}来说t<Sub>i</Sub><t<Sub>i+1</Sub>,并且针对所有i∈{1,2,…}以(算式(I))计算与时间t相关的函数n(t)作为机构的转速的度量。评估单元构造成用于,针对至少一个j∈{1,2,…}以(算式(II))计算与时间t相关的函数m(t)作为机构的转速的梯度的度量；其中,k∈{0,1,…}被选择为适用于：t<Sub>k</Sub>≤t<Sub>j</Sub>-T<t<Sub>k+1</Sub>,T是常数。(The invention relates to a device having a rotatable mechanism, a sensor and an evaluation unit, wherein the mechanism has a number a of markers, wherein the markers pass cyclically through a region detected by the sensor when the mechanism rotates; wherein the sensor is configured for transmitting a signal to the evaluation unit; and wherein the evaluation unit is designed to assign a time t of the transmission signal to each signal i Where i ∈ {0,1, … } and t is t for all i ∈ {0,1, … } i <t i+1 And for all I ∈ {1,2, … } (formula (I)) calculating a function n (I) ((I)) related to time tthe evaluation unit is designed to calculate a function m (t) related to time t as a measure of the gradient of the rotational speed of the machine for at least one j ∈ {1,2, … }, wherein k ∈ {0,1, … } is selected as being suitable for t k ≤t j ‑T<t k+1 And T is a constant.)

1. A device having a rotatable mechanism, a sensor and an evaluation unit; wherein the content of the first and second substances,

the mechanism has a number a of indicia; wherein the content of the first and second substances,

as the mechanism rotates, the indicia cyclically pass through the area detected by the sensor; wherein the content of the first and second substances,

the sensor is configured for transmitting a signal to the evaluation unit; and wherein the one or more of the one,

the evaluation unit is designed to assign a time t of the transmission signal to each signal_iWhere i ∈ {0,1, … } and t is t for all i ∈ {0,1, … }_i<t_i+1And is and

for all i e {1,2, … }, to

Calculating a function n (t) related to time t as a measure of the rotational speed of the mechanism; it is characterized in that the preparation method is characterized in that,

the evaluation unit is configured for: for at least one j e {1,2, … }, to

Calculating a function m (t) related to time t as a measure of the gradient of the rotational speed of the mechanism; wherein the content of the first and second substances,

k ∈ {0,1, … } is selected as appropriate for t_k≤t_j-T<t_k+1Where T is a constant.

2. The apparatus of claim 1, wherein for at least one j' e {1,2, … }, it applies that:

wherein the content of the first and second substances,

n(t_k')<n_minand k' ∈ {0,1, … } is selected as appropriate for t_k'≤t_j'-T<t_k'+1。

3. Device according to any of the preceding claims, characterized in that for at least one point in time t', where t is_j<t'<t_j+1It is applicable that:

m(t')＝m(t_j)。

4. a method for calculating a function n (t) related to time t as a measure of the rotational speed of the rotatable body; wherein the content of the first and second substances,

the mechanism has a number a of indicia; wherein the content of the first and second substances,

as the mechanism rotates, the indicia cyclically pass through the area detected by the sensor; wherein the content of the first and second substances,

the sensor is configured for transmitting a signal to an evaluation unit; wherein the content of the first and second substances,

time t for assigning a transmission signal to each signal_iWhere i ∈ {0,1, … } and t is t for all i ∈ {0,1, … }_i<t_i+1And wherein the first and second end portions of the first and second,

for all i e {1,2, … }, to

Calculating a function n (t) related to time t as a measure of the rotational speed of the mechanism; it is characterized in that the preparation method is characterized in that,

for at least one j e {1,2, … }, to

Calculating a function m (t) related to time t as a measure of the gradient of the rotational speed of the mechanism; wherein the content of the first and second substances,

k ∈ {0,1, … } is selected as appropriate for t_k≤t_j-T<t_k+1Where T is a constant.

Technical Field

The invention relates to a device according to the preamble of claim 1 and a method according to the preamble of claim 4.

Background

The derivative of the measured variable is used in a number of applications. For this, either a simple difference quotient or a digital filter (e.g., Savitzki-Golay filter) is used. As a simple component of the regulation technique, a DT1 cell can be used. Common for all derivative formation is that noise in the base signal results in strong noise of the derivatives. If the derivative is filtered, for example by digital filtering with a large window width or a DT1 cell with a large time constant, a phase shift is created between the input signal and the derivative of the signal.

A common method for detecting the rotational speed is based on detecting and subsequently evaluating the contour of the rotating encoder wheel. For this purpose, for example, a dead time measurement can be used to determine how many teeth pass the sensor per time segment. The time interval between the last two teeth passing the sensor is determined.

The achievable frequency of the sampling of the sensor signal is constantly increasing on the basis of technical developments. Higher sampling frequencies are accompanied by more stringent requirements, but also by better possibilities for more accurate determination of the rotational speed gradient. In many of the previous software applications, the rotational speed is detected every 10 ms. If the step size is reduced, for example to 1ms, then only one tenth of the markings of the encoder wheel passes the sensor in each sampling time section in terms of calculation, without the rotational speed changing. Depending on the rotational speed of the encoder wheel and the number of markings, there are therefore clearly more time steps in which no teeth pass the encoder wheel, i.e. no new rotational speed information is present. However, in the continuous calculation of the rotational speed gradient, it is also necessary to be able to generate a signal at a point in time when no new rotational speed information is available.

Disclosure of Invention

The task of the present invention is to eliminate the disadvantages inherent in the solutions known from the prior art. In particular, a rotational speed gradient with the smallest possible phase shift and low signal noise should be provided. In rotation starting from standstill and in the direction of standstill, the rotational speed gradient is detected as accurately as possible.

This object is achieved by a device according to claim 1 and a method according to claim 4. Preferred developments are contained in the dependent claims.

The device comprises a rotatable mechanism, a sensor and an evaluation unit. The mechanism is free and can be rotated over any angle, in particular over 360 °, without any limitation.

The mechanism has one or more indicia. The number of marks is a. The mechanism may be a code wheel which is fixed in a rotationally fixed manner relative to the shaft or spindle, in particular on the shaft or spindle. The hole disk is suitable, for example, as a coding wheel, which is provided with holes all around. Each two apertures are separated by a tab that extends between the two apertures. The gear wheel is also suitable as a coding wheel, the teeth of which simultaneously serve as markings.

Furthermore, inductive sensors, magnetic field sensors or optoelectronic sensors are suitable as sensors.

The sensor or the area detected by the sensor is oriented towards the mark. As the mechanism rotates, the marks cyclically pass the area of detection. This means that each mark passes exactly once through the area detected by the sensor when the mechanism is rotated 360 °.

The marks pass through the area to be detected by the sensor by each time at least a part of the mark enters the area, passes through the area and comes out of the area again. Upon passing the mark, the sensor generates a signal. The signal may be generated when at least a portion of the mark enters the region, exits the region, or during a period when at least a portion of the mark is in the region.

The signal is transmitted to an evaluation unit. The evaluation unit is designed to assign a time t of the transmission signal to each signal_iWhere i ∈ {0,1, … }, and t is t for all i ∈ {0,1, … }_i<t_i+1And for all i ∈ {1,2, … }, to

A function n (t) related to time t is calculated as a measure of the rotational speed of the mechanism.

Time t of transmission signal_iCorresponding to the point in time when the marker passes the area detected by the sensor.

According to the invention, the evaluation unit is furthermore configured for: for at least one, preferably all, j e {1,2, … }, to

Here, k ∈ {0,1, … } is selected as applicable to t_k≤t_j-T≤t_k+1And T is a constant.

According to the invention, therefore, only the rotational speeds detected at those points in time at which the mark passes through the region detected by the sensor are taken into account for detecting the rotational speed gradient. There is a time interval of at least T between these time points.

In many applications, it is particularly expedient for the rotatable mechanism to decelerate to a standstill and to start moving from standstill. By means of the method according to the invention, gradient values are formed rapidly in the case of a movement starting from rest.

In particular, at slow rotational speeds, a significantly better signal is produced, since the interval between the points in time considered for the formation of the gradient increases. The gradient is then formed with less sensor information in order to obtain the most recent possible value of the gradient. At very low rotational speeds, a gradient is formed over the two last-present sensor signals.

The method according to the invention is characterized in that, at high rotational speeds, i.e. when the sensor signal is present in a sufficiently short time interval, the gradient is not formed over adjacent sampling steps, but rather over a parameterable time constant T which covers a plurality of strokes of the marker (durchgan) through the region detected by the sensor. The noise contribution of the sensor signal into the gradient formation is thus reduced.

Instead of determining the rotational speed gradient as a difference quotient according to the invention, it is conceivable to use digital filtering with variable window widths. This corresponds to a variable number of measured values. Thus, not only the first and last values of adjacent time segments, but also more values are used for determining the gradient.

It is usual to define a minimum rotational speed n to be detected_min. If no mark passes through the area detected by the sensor during the period of time derived therefrom, it is assumed to be stationary. In this case, the rotational speed obtained in theory is still stored for determining the gradient. The value 0 is output as a gradient.

If the rotatable mechanism starts to rotate again after a longer period of standstill, the marking enters the region detected by the sensor on the basis of the rotation. On a stationary basis, the time interval between the detection of the mark and the last mark passing through the area detected by the sensor is very large. This corresponds to a minimum rotational speed n_minThe following rotational speeds. Accordingly, the evaluation unit calculates the value 0 as the rotational speed. The rotational speed gradient is likewise 0.

If the next marking passes through the region detected by the sensor, it is expected that the average rotational speed calculated using the two last detected markings exceeds the minimum rotational speed n_min. If the rotational speed gradient is now formed as a differential quotient, an excessively large value results. To avoid this, it is preferred to improve the evaluation unit such that when n (t)_k')<n_minFor at least one j' ∈ {1,2, … } to

Here, k' ∈ {0,1, … } is selected as appropriate for t_k'≤t_j'-T<t_k'+1。

If the detected rotation speed is at the minimum rotation speed n_minIn the following, not the actual detected rotational speed but the minimum rotational speed n is used_minConsidered for calculating the gradient. A change in the rotational speed gradient is produced which is correspondingly smaller and better reflects the actual rotational speed change.

In a preferred development, at the time t_iThe gap between i ∈ {0,1, … } is interpolated accordingly for at least one point in time t', where t is_j<t'<t_j+1Preferably for all time points t', where t_j<t'<t_j+1According to a further development, the following applies:

m(t')＝m(t_j)。

the method according to the invention is a method as described above, which is carried out by an evaluation unit or a preferred development of the device according to the invention.

Drawings

Preferred embodiments are shown in the drawings. Corresponding reference numerals indicate identical or functionally identical features herein. Specifically, the method comprises the following steps:

fig. 1 shows a first method known from the prior art for determining a gradient;

fig. 2 shows a second method known from the prior art for determining a gradient;

FIG. 3 illustrates a method for calculating a gradient using a time constant;

FIG. 4 shows rotation from rest; and is

Figure 5 shows an improved method.

Detailed Description

Fig. 1 to 5 show the course of the speed function n (t) and the speed gradient m (t) as a function of time t. The assumed course of the rotational speed n (t) corresponding to the calculated gradient m (t) is plotted in dotted lines.

The rotating encoder wheel is sampled by means of a sensor. The sampling takes place at discrete points in time, which are at intervals, i.e. sampling intervals t_sFollowed in succession. Time t_iI ∈ {0,1, … } (at which points in time the markers of the encoder wheel are detected by the sensor) is the sampling interval t_sAn integer multiple of.

For all i ∈ {1,2, … }, the rotational speed function n (t) is calculated from fig. 1 to 4 in the following manner:

the speed gradients m (t) are according to FIG. 1

Although the change of the rotational speed is continuous, the gradient m (t) is at t₃And t₄With pulsed vibration. This can be prevented in that the gradient m (t) is formed as a difference quotient over the last two rotational speed information present in the following manner:

where i ∈ {1,2, … }.

The gradient m (t) shown in fig. 2 is susceptible to noise despite the smooth course of change, and a phase shift occurs, which is dependent on the rotational speed. This problem can be solved in that the gradient m (T) is formed over a parametrizable time constant T as shown in fig. 3. Here, for j ∈ {1,2, … }, the gradient m (t) is calculated as:

where k ∈ {0,1, … } is selected to apply to t_k≤t_j-T≤t_k+1Where T is a constant.

Fig. 4 shows the course of the speed function n (t) in the encoder wheel, which starts from rest. The dashed line shows the actual speed profile. Based on at a point in time t₁And t₂Long rest time in between, at a point in time t₃A rotational speed well below the actual rotational speed is calculated. This in turn results in a gradient m (t) at t₃≤t<t₄Forms a peak within the range of (a).

As shown in fig. 5, such a peak value can be set by presetting a minimum rotational speed n_minAnd (4) eliminating. If at the time t the rotational speed n (t) calculated from the sensor signal is less than the minimum rotational speed n_minThen the minimum rotational speed n_minInstead of the calculated rotational speed n (t), this is taken into account for determining the gradient. For t₃≤t<t₄M (t) is thus calculated as:

t is here parameterized as T for simplicity_s。