KR20170052509A - Method for prediction of a period between two signal edges of a rotational speed sensor signal - Google Patents

Abstract

The present invention relates to a method for predicting a time period between two signal edges of a rotational speed sensor signal. The method includes: providing a signal (S) for determining the rotational speed of a combustion engine (100) by scanning an encoder wheel (120) of the combustion engine (100); determining a first reference period and a second reference period between active reference signal edges, respectively; and estimating the first period in consideration of the second period. The method is suitable for deriving rotational speed information and/or crankshaft angle information.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a method for predicting a period between two signal edges of a rotational speed sensor signal,

The present invention relates to a method for predicting a time period between two signal edges of a rotational speed sensor signal, and to a computation unit and a computer program for performing such a method.

An in-car combustion engine generally has a so-called encoder wheel, and the rotational speed or crankshaft angle of the combustion engine can be detected by the encoder wheel. To this end, the encoder wheel is scanned by the rotational speed sensor. Since the encoder wheels have toothed gears with a constant angular separation from each other, the rotational speed sensor provides a signal with alternating rising and falling signal edges. By evaluating these two types of signal edges, so-called active signal edges, rotational speed information or crankshaft angle information can be provided.

In accordance with the present invention, a method for predicting a time period between two signal edges of a rotational speed sensor signal, having the features of the independent patent claims, and a computation unit and a computer program for performing such a method are proposed. The preferred embodiments are subject of the dependent claims and the following detailed description.

The method according to the invention with the features according to claim 1 is characterized in that in estimating a first period between these signal edges in a state in which at least one of two successive active signal edges of one signal of the rotational speed sensor has not yet been detected .

As already mentioned at the introduction, rotational speed information and / or crankshaft angle information for the combustion engine may be provided by the rotational speed sensor and the encoder wheel. However, such information is generally only available to the last collected, ie, the last measured or evaluated signal edge. For example, a certain number of equidistant angular tick marks are inserted between two consecutive active signal edges to provide more detailed information about the rotational speed or crankshaft angle than by the active signal edge . However, for information after the last collected active signal edge, for example, it is necessary to know the period between the last active signal edge and the next active signal edge that has not yet been detected.

As already mentioned in the introduction, the rotational speed sensor can be fixedly mounted on the crankshaft of the combustion engine and can include a series of upward and downward movements associated with the teeth of the encoder wheel passing near the rotational speed sensor upon rotation of the combustion engine Signal edge. In general, only a falling edge or a rising edge is observed, in which case these edges are referred to as active signal edges. The occurrence of such an edge is related to the position of the associated encoder wheel teeth. An angular counter can also be implemented that continuously counts a predetermined number of angular check indications between the two active signal edges to obtain subdivided angular information or rotational velocity information between the edges. The spacing between the two active signal edges is also referred to below as "incremental" or "duration ". At this time, the frequency of occurrence of the angle check marks corresponds to the number of angle check marks divided by the increment duration or duration to be accurately known only at the time of the next active signal edge. Since the frequency is newly set as each new signal edge for subsequent increments, it would be desirable if the time until the next signal edge arrives can be predicted as accurately as possible. In this case, a crankshaft signal memory may be accessed that stores a history of past collected periods over a predetermined time interval, until the oldest value is overwritten with the new value again. Within the recorded range there is included a specific characteristic torque variation repeated for each combustion cycle, for example, a compression pattern upon compression of air or air-fuel mixture in the cylinder, Can be considered for improving the quality of prediction.

For estimation in the sense of predicting such a period, it is now advantageous to use situations in which a series of sawtooth periods (periods) between different cylinders or combustion periods of the combustion engine have high similarities, The sawtooth period (period) can be mutually mapped by linear rescaling.

Thus, in the case of the upcoming period? T _i , for example, the immediately preceding period? T _{i -1} , and the associated reference periods? T _{i -m} and? T _{i -m- 1} ), an estimated value is indicated, using the ratio of immediately subsequent reference periods. Therefore, the following expression is obtained.

Where m corresponds to the tooth spacing between the cylinders or the tooth spacing during one combustion period. Wherein the cylinder spacing means the angular spacing of the ignition points of the individual cylinders of the combustion engine, in which case one combustion period forms the product of the cylinder spacing and the number of cylinders. For example, in a four-cylinder engine, when one combustion period is 720 °, the cylinder spacing is 180 °, which again corresponds to two full rotations of the crankshaft.

However, in this case, there arises a problem that the rotational speed fluctuation within the reference range in the past, that is, the one-cylinder interval or the reference periods subsequent to the one combustion period, can not always be devoted in the future. In the case of a so-called compression torque which depends only on the crankshaft angle, but not on the rotational speed at all, the above-mentioned possibility of exclusive is less guaranteed as the rotational speed difference between the predicted range and the reference range is larger.

Further, only the immediately preceding period history may be referred to, for example, only the obesity of the immediately preceding two periods can be referred to. Thus, for example, the following equation is obtained.

In this case, it should be premised that the rotational speed tendency (rise or fall) that existed in the preceding period continues in the future. If not (for example, when vibration occurs on the crankshaft in the vicinity of the turning point or on the first combustion cycle), reliable estimation of the next period is impossible.

]에 의해 증분 지속 시간 또는 기간의 단기 변동이 야기된다는 점이 전제된다. 이것이 가능한 이유는, 제1 근사에서의 가정이 압축 토크의 회전 속도 의존성과 관련한 실제 상황들에 상당히 부합하기 때문이다. 일 실린더 안에 포함된 가스량의 압축을 위해서는 회전 속도와 무관하게 항상 동일한 힘이 필요하기 때문에, 상기와 같은 사실은 특히 이를 통해 야기되는 가속 토크 또는 제동 토크에도 적용된다. 이 경우, 하기의 식,Now, in the proposed method, within the reference range, that is, behind the predicted range, several times the cylinder spacing, and within the predicted range, the same angle dependent torque [

] Causes a short-term variation of the incremental duration or duration. This is possible because the assumptions in the first approximation substantially correspond to the actual situations relating to the rotational speed dependence of the compression torque. Since the same amount of force is always required to compress the amount of gas contained in one cylinder, regardless of the rotational speed, this fact applies particularly to the acceleration torque or braking torque caused thereby. In this case,

Where J is the moment of inertia of the combustion engine, and according to the variable separation method,

Is applied, where? Is the angular velocity. By integrating the above equations,

.

및

)를 갖는 예측 범위 내 및 기준 범위 내에서 2개의 이웃하는 기간과 관계시키면, 전술한 가정에 상응하게 상기 두 범위의 토크 작용이 동일해진다. 따라서, 하기의 식,The equation can be expressed as an angle corresponding to active signal edges or reference signal edges

And

) And the two neighboring periods within the reference range, the torque acts of the two ranges are made corresponding to the above-described assumption. Therefore,

Where DELTA phi is the uniform angular spacing between two consecutive active signal edges or the spacing of two teeth of the encoder wheel, DELTA _ti is the first period, DELTA _{ti- 1} is the second period, ? T _{i -m} is one reference period, and? T _{i -m-1} is a second reference period. Since the angular interval DELTA phi is constant,

.

That is, the first reference period (Δt _{i -m)} with the second reference period torque action is the first time period (Δt _i) is required to cause the movement of the encoder wheel variation corresponding to the difference between (Δt _{i -m-1)} And the second period (? T _i-1 ).

The advantage of the proposed method over the previously mentioned method is that a more realistic prediction can be realized, for example in the case of a rotational speed increase or a rotational speed reduction, that is, when the reference range and the prediction range do not lie at the same average rotational speed It can be.

In the proposed method, for example, during engine acceleration, the absolute value of the difference between the first reference period (Δt _im ) and the second reference period (Δt _{i -m-1} ) decreases in the proposed transformation, ,

Is applied.

At the time of engine deceleration in the proposed method, the absolute value of the difference between the first reference period (? T _{i -m} ) and the second reference period (? T _im-1 ) at the time of conversion increases,

Is applied.

On the other hand, in the case of the above-mentioned method, the relative proportions of the periods are the same, that is,

Is applied.

In other words, the proposed method considers that the rotational speed variation decreases in inverse proportion to the rotational speed due to the periodic torque irrespective of the rotational speed. On the contrary, the above-mentioned method misrepresents the opposite proportional principle, that is, the rotation speed variation is directly proportional to the rotation speed.

Preferably, the first period is defined by the following formula:

Or is calculated from the Taylor series expansion of the formula, so that the torque action is devoted to the first period and the second period.

Solving this equation yields the formula described above for the first period. In this way, very simply the first period is formed and determined as a function of the second period and as a function of the first and second reference periods.

In this case, the formula as Taylor series expansion is preferably considered, and the Taylor series expansion is used up to at least the first order, in particular at least the second order. In the above formula, the equation can be simplified by a Taylor series expansion for simpler and faster computation, because in the calculation unit, for example a control unit, a root can be generated, which can only be calculated by a very complicated calculation have. In this connection,

Is assumed to be applied, which has been met at a very good approximation here. In this case,

By the following equation,

The first period can be obtained from the second Taylor series expansion as the second period.

Alternatively, it is also desirable to determine the square root of the formula in consideration of the characteristic curve and / or the look-up table. In this way too, complicated calculations in the calculation unit can be avoided.

It is preferred that the first period is limited by the signal edge which has not yet been detected and by the signal edge which has already been detected or determined, or alternatively by the two signal edges which have not yet detected the first period. Here, the detected signal edge means a signal edge that already exists in the signal detected by the rotational speed sensor, while the determined signal edge may mean an estimated signal edge, in particular according to the method proposed here . That is, in the above-described first alternative with the already detected signal edge, a first period elapsing from the already detected second signal edge to the next signal edge to be detected further can be estimated. Thus, very accurate estimation or prediction is possible with only a short time interval at which the signal edge to be detected is to be detected. In this case, in the above-described first alternative with the signal edge already determined, a first period which has already been determined, i.e. elapsing from the estimated second signal edge to the next signal edge to be detected further, can be estimated . In this way, for example, by applying this method several times, it becomes possible to predict a much earlier period. In the above-mentioned second alternative, for example, a much later later period of time can be directly estimated.

Preferably, the periods are each limited by two successive signal edges, the reference periods being each limited by two successive reference signal edges. In this way, very accurate prediction is possible with only minor differences between the two periods or the reference period.

Preferably, in consideration of the estimated first period, rotational speed information and / or crankshaft angle information between two signal edges that limit the first period is derived. As already mentioned in the introduction, estimating or predicting corresponding periods is required for rotational speed information and / or crankshaft angle information for a range of signal edges that have not yet been detected. The proposed method is well suited for deriving rotational speed information and / or crankshaft angle information, since the proposed method allows for very accurate estimation of the period.

It is preferred that angle clock, angle-time conversion, rotational speed information for rotational speed prediction and / or crankshaft angle information are used for starter control and / or injection programming. Such an application is particularly an example applied to an automobile which requires very precise rotational speed information and / or crankshaft angle information which can be provided by the proposed method. Therefore, as a possible method for estimating the periods of the plurality of future time points described above for starter control, which generally requires accurate stopping positions of the crankshaft, a crankshaft angle at which the crankshaft is stopped at the time of costing of the combustion engine Can be accurately determined. In injection programming, for example, very precise temporal adjustment of these injection processes can be carried out as desired, especially if future injection processes are carried out at very short intervals of time.

A control unit of a calculation unit, for example an automobile, according to the present invention is designed to perform the method according to the invention, in particular programatically.

It is also desirable to implement this method in the form of a computer program, since this approach results in particularly low costs, especially if the running-side control device is used for other tasks anyway. Data carriers suitable for providing computer programs are, in particular, magnetic memories such as hard disks, flash memories, EEPROMs, DVDs, optical memories, and electrical memories. It is also possible to download programs via a computer network (Internet, intranet, etc.).

Further advantages and embodiments of the present invention refer to the detailed description and the attached drawings.

The present invention is schematically illustrated in the drawings based on one embodiment, and is described below with reference to the drawings.

Figs. 1A and 1B schematically show two different views of an encoder wheel with a rotation speed sensor. Fig.
2 is a graph showing a signal waveform of the rotation speed sensor.
3 is a graph depicting a time period and a reference period as can be used in one preferred embodiment of the method according to the present invention.
FIGS. 4A and 4B are graphs showing the results of comparing the ratio of the estimated period to the actual period in a preferred embodiment of the method and the method according to the present invention, unless the present invention is applied.

Figure 1 a schematically shows a combustion engine 100 with a crankshaft 110. In the crankshaft 110, an encoder wheel 120 fixedly connected to rotate with the crankshaft 110 is disposed. Also shown is an associated rotational speed sensor 130 that generates a signal based on the scanning of the encoder wheel 120. In this case, the rotational speed sensor 130 may be fixedly disposed at an appropriate position, for example.

Also shown is a calculation unit 140 formed as a control device designed to control the combustion engine 110. Further, the calculation unit 140 can collect signals generated by the rotation speed sensor 130. [

의 간격으로 배치되며, 다시 말해 60개의 톱니가 제공된다. 본 도면에 도시된 예에서는 편의상 톱니 수가 더 적게 도시되어 있다.1B, a front view of the encoder wheel 120 and the rotational speed sensor 130 is shown. In this figure, the teeth 121 formed at the outer edge of the encoder wheel are shown. Generally,

In other words, 60 teeth are provided. In the example shown in the figure, the number of teeth is shown for convenience.

The rotational speed sensor 130 may, for example, have a coil, i. E. May be formed as an inductive sensor. In this way, a signal is induced in the coil of the rotational speed sensor 130, which is formed by the teeth 121 or by a gap lying between the two teeth, respectively.

Now, in Fig. 2, the signal S provided by the rotational speed sensor when scanning the encoder wheel is shown. In this regard, the signal S is shown for time t.

At this time, the signal S shows a rectangular waveform with rising and falling signal edges. Such rising and falling signal edges are generated by the rotation of the teeth of the encoder wheel through the vicinity of the rotational speed sensor. To determine the rotational speed or crankshaft angle of the combustion engine, the time interval of two successive rising signal edges or falling signal edges is generally detected. In this case, the rotation speed can be calculated very simply by the constant angle DELTA phi.

In this case, for example, the rising signal edge is selected as the active signal edge (F), that is, the signal edge used to determine the rotational speed. At this time, the time interval of the two consecutive active signal edges F indicates the duration or the incremental duration? T. In the illustrated example, the signal S up to the time point t ₀ has already been detected, which is indicated by a solid line.

3, there is shown a time period and a reference period as can be used in one preferred embodiment of the method according to the present invention. In this regard, the period? T is shown for time t. Where V represents the waveform of the period DELTA t over time t for two different rotational speed ranges. The left portion shows the waveform at the higher rotational speed correspondingly with the shorter period and the right portion shows the waveform at the lower rotational speed and correspondingly longer period.

Further, it can be seen from the waveform (V) that the period is not constant but fluctuates even within one rotation speed range. The reason is that even if the rotational speed is constant on average over one combustion cycle of the combustion engine, the rotational speed changes due to the different torques that the offset cylinder applies to the crankshaft.

Subsequently, a first period (? T _i ) and a second period (? T _{i - 1} ) are shown on the right side of the waveform (V). In the left part of the waveform V, a first reference period? T _{i -m} and a second reference period? T _im-1 are shown. At this time, the first period may be a first period of time (Δt _i), in other words the prior limit the active edge of the signal a first time period has not yet been detected by the speed sensor (Δt _i) in Fig. 2 e.g. .

)만큼 오프셋되어 있다. 이러한 방식으로, 크랭크 샤프트에 작용하는 각도 의존적 토크가 두 가지 경우 모두 동일한 점이 보장된다. 이는, 2개의 기준 기간에 대한 기준 토크[

] 및 2개의 기간에 대한 예측 토크[

]로써 도시되어 있다.At this time, the period and the reference period are, for example, a cylinder interval

). In this way, the angle-dependent torque acting on the crankshaft is guaranteed in both cases. This is because the reference torque [

] And the predicted torque for the two periods [

As shown in FIG.

Thus, the first period is calculated or calculated from the second period and the first and second reference periods in accordance with the above-described formulas.

4A and 4B, in a preferred embodiment of the method according to the invention and the method not according to the invention, the ratio of the estimated period to the actual period over time (t) is shown respectively have.

In the ideal case, the ratio should take a value of 1 constantly. At this time, as the reference period, for example, the periods derived from the signal memory of the rotational speed sensor, which may be in the calculation unit, are referred to. In this case, the reference period is, for example, one past combustion period.

The proposed method (see FIG. 4B) is activated only at the point of use of the reference periods, in this case from point of view t ₁ . It can be seen from the figure that in the case of a method not according to the invention (see FIG. 4a), after activation, in this case again at time t ₁ , Is much larger than the case of.

Claims (11)

At least one of two consecutive active signal edges F of one signal S of the rotational speed sensor 130 has not yet been detected and a method for estimating a first period? T _i between these signal edges As a result,
The rotational speed sensor 130 provides a signal S for determining the rotational speed of the combustion engine 100 by scanning the encoder wheel 120 of the combustion engine 100,
A second period (? T _{i - 1} ) between the two consecutive active signal edges (F) is determined, and at least one of the two signal edges does not limit the first period (? T _i )
A first reference period (? T _{i -m} ) and a second reference period (? T _im-1 ) are determined between two successive active reference signal edges,
Each of the reference signal edges and the corresponding signal edge is spaced at one cylinder spacing

), ≪ / RTI >
A first reference period of time (Δt _{i -m)} with the second reference period (Δt _{i -m-1)} a torque acts required to cause a change operation of the encoder wheel 120 corresponding to the difference of the first period (Δt between _i ) and the second period (? t _{i - 1} ), whereby the first period (? t _i ) is estimated taking into account the second period (? t _{i -1} ). The method of claim 1, wherein the first period (? T _i ) is calculated from a second period (? T _{i - 1} )

, The torque action is dedicated to the first period (t _i ) and the second period (t _{i - 1} ), wherein Δt _i represents the first period and Δt _i is the first period _{i -1} represents a second period,? t _{i -m} represents a first reference period, and? t _im-1 represents a second reference period. 3. The method of claim 2, wherein the first period (? T _i ) is calculated according to a Taylor series expansion of the formula from a second period (? T _{i - 1} ), and wherein the Taylor series expansion is used up to at least a first order. The method of claim 2, wherein the first period of time (Δt _i) is the second _period is calculated according to the formula from (Δt _{i 1),} which is the square root appears in the formula is determined in consideration of the characteristic curve and / or look-up table, Estimation method. The method according to any one of claims 1 to 4, wherein the first period (? T _i ) is limited by the signal edge not yet detected and the signal edge already detected, or the first period (? T _i ) And is limited by two signal edges that are not equal. The method according to any one of claims 1 to 4, wherein the period (? T _i ,? T _{i - 1} ) is limited by two immediately successive signal edges and the reference period (? T _{i -m} ,? T _{i -m-1} ) is limited by two successive reference signal edges. Any one of claims 1 to A method according to any one of claim 4, wherein estimating the first period (Δt _i) the way, the first period of time (Δt _i) the limited two signals rotational speed and / or crankshaft between the edge under consideration Angle information is derived. 8. The method of claim 7, wherein an angular clock, angular-time conversion, rotational speed information for rotational speed prediction and / or crankshaft angle information is used for starter control and / or injection programming. A calculation unit (140) designed to perform the method according to any one of claims 1 to 4. A computer program stored on a machine-readable storage medium, the computer program being executable on the computing unit (140) to cause the computing unit (140) to perform the method according to any one of claims 1 to 4. 12. A machine-readable storage medium having stored thereon a computer program according to claim 10.

Images