JP4071963B2 - Method and apparatus for calculating engine speed according to reality - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention uses a sensor that scans a gear connected to the crankshaft of the engine, the gear having a plurality of teeth spaced at regular intervals on the surface, as the teeth pass the sensor. The signal corresponds to the surface of the gear, the signal is evaluated by the engine control device to calculate the rotation speed, and the rotation speed is calculated from the set time interval of the side edges of the signal. The present invention relates to a method for calculating a rotational speed in accordance with reality.
[0002]
[Prior art]
The engine speed is a very important amount for engine control. This is because various engine operation amounts are calculated from the rotational speed. Engine operation amount in today's engines, and the amount sent from the engine control unit, which is calculated in dependence on the rotational speed information by the engine control unit, various different time patterns, for example 10 ms, 20 ms, in each 50 ms, predetermined as well in pattern angles of the synchronization, it is calculated from the engine speed realistic. Pattern to synchronize with predetermined angles of this is once per combustion is determined four times in the operating cycle the angle of the crank shaft is 720 ° with respect to namely 4-cylinder engine. The interval from the occurrence of synchronization with the predetermined angle to the next generation of synchronization with the predetermined angle is referred to as SYNCHRO in this specification. That is, when one SYNCHRO is completed , the crankshaft rotates 180 ° in the case of a four-cylinder engine . The interval at which such synchronization is taken is also referred to as a normal segment interval.
[0003]
The engine speed nmot is obtained from the angle information. The engine speed nmot is calculated at the start of SYNCHRO, for example. Here, the engine speed is calculated from the time interval until the start of the previous SYNCHRO. Determining the engine speed in this way is realistic for each start of SYNCHRO, but is averaged over a 180 ° crankshaft angle (in a four cylinder engine).
[0004]
This value thus determined for the engine speed is used for all calculations requiring the engine speed until a new engine speed is calculated again at the start of the next SYNCHRO. used. When the engine speed is low, for example, in the engine starting stage, the actual engine speed changes greatly. Therefore, the amount nmot calculated at the start of SYNCHRO is not realistic enough to control the engine very accurately.
[0005]
[Problems to be solved by the invention]
An object of the present invention is to provide a method for calculating an engine speed in accordance with reality as follows. That is, at the start stage of the engine having a small number of revolutions, an actual engine revolution number considered for controlling the engine is formed. At this time, the tooth time should not be additionally calculated in order to save the resources of the processor or the computer.
[0006]
[Means for Solving the Problems]
This task is to calculate the rotation speed by the first method for a rotation speed larger than the set threshold value, and to rotate by the second method for the rotation speed below this threshold value or the second threshold value. It is solved by calculating the number.
[0007]
DETAILED DESCRIPTION OF THE INVENTION
In today's engines, the engine speed is calculated using a gear connected to the crankshaft, which is usually scanned by a sensor. The gear has 60 minus 2 teeth on the surface. Here, the two missing teeth are arranged side by side, forming a so-called reference mark, which is required to determine the absolute position of the gear and thus the crankshaft. As the gear rotates, the teeth pass through a fixed sensor. A sensor, for example an inductive sensor, thus forms an output signal, the course of which depends on the surface of the gear. This output signal exists as a square signal after being properly processed. This square signal is evaluated by the engine controller to determine the angular position of the crankshaft and to calculate the engine speed. Such a system is described, for example, in EP-OS 44 34 833.
[0008]
In evaluating such rotational speed-dependent square signals, the controller micro-controller mechanism stores the time points of each side edge or settable side edge of the square signal in the drum memory. In order to not overburden the controller microcontroller and to limit the effects of speed fluctuations, current systems do not calculate the speed from each side edge of the sensor square signal. The current method is described below.
[0009]
The method according to the invention for calculating a realistic engine speed having the features of claim 1 has the following features. In other words, even if the engine speed is low, a value nmot that matches the actual engine speed can be obtained, and this actual value can be used for another calculation of the amount that depends on the engine speed. This is particularly advantageous because the fluctuations in the speed do not have an adverse effect on the calculation of other engine operating quantities based directly or indirectly on the engine speed, especially in the starting phase. is there. Therefore, all time patterns that require the engine speed as a calculation amount can use values of the engine speed nmot that are clearly more realistic than the current system.
[0010]
This advantage is achieved by using realistic angle information to calculate the rotational speed. It is particularly advantageous that in current engines, the rotational speed information, which is usually present in the drum memory or ring memory of the control unit, is used to calculate the rotational speed and is necessary for calculating the rotational speed. The corrections taken are considered together. That is, by using a suitable evaluation algorithm, only the rotational speed information already present in the drum memory anyway is better evaluated in an advantageous manner.
[0011]
It is particularly advantageous to switch between the previously used calculation algorithm, which can be left at a relatively high rotational speed, and the new algorithm used for a low rotational speed.
[0012]
Further advantages of the invention are achieved by the features of the dependent claims.
[0013]
Embodiments of the present invention will be described below in detail with reference to the drawings.
[0014]
【Example】
FIG. 1 shows an embodiment for detecting the engine speed, which can also be used to implement the method according to the invention. A gear 1 connected to the crankshaft 6, for example a gear having 60 minus 2 teeth, is scanned by a fixed sensor 5, for example an induction sensor. In order to determine the priority position, the gear 1 connected and fixed to the crankshaft 6 has a reference mark 2, for example where two teeth are missing. The tooth of the gear 1 is represented by 3, and the reference symbol 4 is given to the side edge of the tooth. As the gear 1 rotates, the teeth pass the sensor and a voltage is induced in the sensor 5a. Using a suitable signal shaping circuit 5b, a pulse train S1 is formed as the output signal of the sensor 5a, which reflects the surface of the gear 1.
[0015]
The output signal S1 of the signal shaping circuit 5b is supplied to the processor 17 that performs evaluation. Here derives synchronization signals using an interrupt device 8 from the signal S1, the pulse interval is equidistant with respect to angle, also depends on the number of cylinders of the engine. Further, a time value depending on the rotation speed is formed from the signal S1 in the time circuit 9, for example, and this time value is supplied via the switch 10 to the ring memory 11 which is a component of the processor 17, for example. Here, the ring memory 11 has a configurable number of memory elements. Re-storing is performed by controlling the switch 14 respectively.
[0016]
From the contents of the settable ring memory elements supplied via the switches 15 and 16 of the calculation unit 18 in the processor 17, the calculation unit 18 calculates a realistic engine speed in a settable way. Under certain settable conditions, the calculation unit 18 can calculate the rotational speed even from the time interval of the synchronization signal pulses.
[0017]
Using the apparatus shown in FIG. 1, the evaluation method according to the present invention described below can be realized. In order to understand this method, first, the signal course S (t) shown in FIG. 2 will be described. FIG. 2 shows a calculation pattern on the first time axis by taking a 4-cylinder engine rotating at 200 revolutions per minute as an example. In such engines, a synchronization signal is generated every time the crankshaft rotates 180 °. The synchronizing signals that follow each other are represented by A, B, C, and D in FIG. In this example, the time interval between two sync signals or two sync pulses, for example, AB is 150 ms. Between the synchronization signals, pulses generated by the side edges of the sensor signal are shown. The interval between the two synchronization signals corresponds to 30 teeth , and the region where the teeth are missing (reference mark) is not shown. For example, in this region, pulses can be generated by simulating so-called teeth.
[0018]
Depending on the mechanism in the microcontroller used, each time point of the tooth pulse is stored in the drum memory. However, for normal rotation speed calculation, only the entry at the start of the actual SYNCHR and the start of the preceding SYNCHR is evaluated.
[0019]
For example, two calculation patterns, a 10 ms pattern and a 50 ms pattern are represented on the second time axis. Here 50ms pattern is are numbered from _50/1 to _50/9.
[0020]
When each of the synchronization signals A, B, C and D is generated , the engine speed nmot is newly calculated. For example, in the synchronization signal B, the rotation speed nmot, B is calculated by evaluating the time elapsed between AB. This rotational speed nmot, B is used for the subsequent calculation in 50 ms, and is also used for the calculation every 10 ms between BCs. In the synchronization signal C, the engine speed nmot, C is newly calculated and thereafter used for all subsequent time patterns. It can be seen from FIG. 2 that when the engine speed changes drastically, it is calculated using the old value of nmot too long. This is because, despite the fact that there are already a large number of new sensor signal pulses, and thus rotation speed information, they are still not calculated anew.
[0021]
Similarly, it can be seen from FIG. 2 that the number of revolutions nmot calculated from the time interval between the two synchronization signals represents the average number of revolutions for the last 30 teeth or the corresponding square signal. If the rotational speed varies, whether the rotational speed increases or decreases, the quantity nmot averaged over 30 teeth is approximately the same as the actual rotational speed in the middle between the two synchronization signals .
[0022]
As described above, the rotation speed nmot, B is calculated when the synchronization signal B is generated. When the rotation speed fluctuates, the rotation speed nmot, B substantially corresponds to the actual rotation speed for the time point 50 (2). This rotational speed is used until a new amount nmot, C is supplied in the synchronization signal C. In the pattern 50 (6), the rotation speed information is about 200 ms old. The same applies to patterns every 10 ms. In the 10 ms pattern before the synchronization signal C, the rotational speed information nmot, B used is even older than 200 ms.
[0023]
FIG. 3 shows another signal course S (t) or a specific calculation pattern taking a four-cylinder engine at 1000 revolutions per minute as an example. The synchronization signal interval, for example the interval between ABs, is 30 ms in this case. Thus, it can be seen that when the rotational speed is relatively high, the amount nmot calculated at the start of SYNCHRO is not very realistic. That is, it can be seen that the quantity nmot, B is about 40 ms older in the calculation pattern 50 (2). Using the method according to the present invention for calculating the actual rotational speed, for example, when the rotational speed is low, the actual rotational speed is calculated, and this rotational speed is then controlled based on the rotational speed information. Each can be used for calculations performed by the device.
[0024]
According to the invention, different possibilities are used for the time of calculation and for the calculation method for the actual rotational speed. This will be described in detail below.
[0025]
In the method according to the invention, the rotational speed nmot is no longer calculated using the conventional algorithm at the start of SYNCHRO under certain settable conditions, for example when the rotational speed is low, but a new Calculated using a calculation algorithm. The possible time points for the calculation are: That is,
a) Calculate only for example in a time pattern every 10 ms, or b) Calculate only at the start of SYNCHRO, or c) Calculate for example in a 10 ms time pattern, and also at the start of SYNCHRO.
[0026]
The calculation of the rotation speed nmot by the means listed from a to c can be associated with a predetermined condition, for example. Here, the conditions can be selected depending on the engine speed. In this case, for example, when the engine speed is low, the condition is calculated by the first means, and the engine speed is a threshold value, sometimes with hysteresis. As long as it is greater than the threshold, it is calculated by another means. For example, the usual measures are then taken again, and the rotation speed is calculated from the time interval between successive synchronization signals . If necessary, further conditions can be taken into account.
[0027]
The switching condition for calculating the rotational speed can be calculated in the same time pattern as the rotational speed calculation or in a different time pattern.
[0028]
When calculating the engine speed in the time pattern, it is not necessary to uniquely determine the tooth time. Rather, information in the drum memory about the last time of the teeth or the last time of the configurable signal interval is used. In addition to the drum memory with tooth time, there is a pointer to the location in the drum memory where subsequent entries are to be stored. Both pointers are evaluated. The pointer carries information about the last entry in the drum memory. In considering another previous entry in the drum memory, if the beginning of the drum memory is reached, the entry before it occurs at the end of the drum memory. If those entries have not yet been overwritten, they still exist and can be read at that location.
[0029]
In evaluating drum memory entries, one or more algorithms described below can be applied. All algorithms are common in that the number of revolutions is calculated from the time interval between two entries in the drum memory and the angular interval between these entries.
[0030]
In the method according to the invention, the time between the last two teeth is evaluated. For this, the last two entries in the drum memory are evaluated. In another method, the time between the last tooth and the nth previous tooth is evaluated. For this purpose, the last entry and the entry before the n position (where n is 2 to 29 for a 30 tooth segment length) are evaluated.
[0031]
Another evaluation method uses the time between the last tooth and the previous tooth by one segment length. In this case, the last entry in the drum memory and the previous entry by one segment length are evaluated. Older entries at the same location in the drum memory, however, have been overwritten by the last entry. When using the second drum memory, access to older entries is possible. This second drum memory can always be written when detecting the rotational speed by alternately writing to one of the two drum memories, or the second drum memory can be written from a time pattern. Can do. The advantage of this calculation method is that the influence of unevenness of the engine course is eliminated since the engine speed is averaged over the SYNCHRO length at each time point. In this algorithm, since the rotational speed is also averaged over the segment length here, realistic rotational speed information is not provided for SYNCHRO.
[0032]
In another method according to the invention, the time between settable fixed teeth is evaluated in order to determine a realistic rotation speed. For example, between the tooth 2 and the tooth 6 or between the tooth 17 and the tooth 21, it is evaluated in each case which difference is more current. This means takes into account unevenness of the engine course. In the compression stage of the engine, the rotational speed is smaller than the average value over the segment length or synchronization signal interval, whereas it is larger in the combustion stage. If the rotational speed is represented in a diagram over the time when the crankshaft is 180 °, it is possible to observe the rotational speed course having substantially the same characteristics as the curve drawn by the solid line in FIG.
[0033]
FIG. 4 shows the relative engine speed as a function of time between two synchronization signal marks or synchronization signal pulses. Unevenness in the number of rotations can be identified from the solid line. Here, there is a region where the engine rotational speed nmot is larger than the average rotational speed or the average value of the rotational speeds. These regions are represented by B1 and B5. The region (B3) has a nmot smaller than the average rotational speed, and the region (B2, B4) has a course of nmot substantially the same as the average rotational speed. If only this latter region is evaluated, the speed determined in this way is approximately equal to the average value of the speed. In FIG. 4, a dashed arrow is shown, which represents which 10 ms pattern in either of the two possible tooth time information B2 or B4 is to be evaluated.
[0034]
In another method according to the invention, the corrected time between the last two teeth is used to calculate the engine speed. Here, the corrected time between the last two teeth is evaluated and this time is corrected by a factor representing the non-uniformity of the engine rotation. FIG. 4 shows the relative engine speed as a function of time as a curve drawn with a solid line. When the engine speed is calculated at a time when the relative engine speed is 1.1 times the average value due to the system, it is then divided by the factor 1.1 or even multiplied by the factor 1 / 1.1 Therefore, the corrected average engine speed can be calculated. That is, the side edges of all teeth are multiplied by the value from the characteristic curve depending on the number of teeth from the beginning of SYNCHRO.
[0035]
Here, the characteristic curve includes the reciprocal of the function shown in FIG. The average speed is therefore calculated by the following formula:
[0036]
nmot, average = nmot, two last teeth * coefficients where the coefficient is a function of the number of the corresponding tooth. The coefficient as a function of the tooth number can be calculated by first defining a constant 1.0. Here, if the engine speed nmot is calculated in the time pattern of 10 ms over the segment length or over the interval between the two synchronization signals , a speed course corresponding to the curve drawn by the solid line in FIG. 4 occurs. From there, a coefficient as a function of tooth number can be calculated. The rotational values determined by the method according to the invention can additionally be further averaged and / or filtered.
[0037]
It is also possible to limit the change. In yet another evaluation method, the results are further validated. Here, the rotation speed nmot, 10 ms calculated in the time pattern has, for example, a maximum deviation with respect to the quantities nmot, SYNCHRO calculated in SYNCHRO. If the number of revolutions nmot, 10 ms departs from the set allowable range around nmot, SYNCHRO, a settable restriction is performed. The allowable width can be expressed as follows, for example.
[0038]
0.5 nmot, SYNCHRO <nmot, 10 ms ≦ 1.5 nmot, SYNCHRO
Here, if necessary, in order to avoid a drastic change in the rotational speed, the allowable width can be always reduced slightly before reaching the switching rotational speed. Therefore, from the calculation of the new rotation speed described here, the interval between two synchronization signal pulses is evaluated in order to determine the rotation speed as described above. Transition occurs.
[0039]
Each of the above-described methods for calculating the engine speed is more realistic than the method for calculating the engine speed from the interval between the two synchronization signals . Supply results. What is important is that when the rotational speed is low, the rotational speed is calculated or calculated in a fast time pattern. On the other hand, when the rotational speed is relatively high, for example, the rotational speed is higher than a settable rotational speed threshold. If it is high, it is switched to the calculation of the number of rotations in the synchronization signal interval.
[Brief description of the drawings]
FIG. 1 shows in detail a rotational speed detection system capable of carrying out the method according to the invention.
FIG. 2 shows a signal course of 200 revolutions per minute.
FIG. 3 shows a signal course of 1000 revolutions per minute.
FIG. 4 shows the relative engine speed as a function of time.
[Explanation of symbols]
1 gear, 2 related mark, 3 teeth, 4 tooth side edges, 5a sensor, 5b signal shaping circuit, 6 crankshaft, 8 interrupt circuit, 9 time circuit, 10 switch, 11 ring memory, 14, 15, 16 switch, 17 processors, 18 computing units

Claims (12)

A method for calculating the engine speed in line with reality,
Using a sensor that scans the gears connected to the crankshaft of the engine,
The gear has a plurality of teeth spaced regularly on the surface;
The teeth form a signal as they pass through the sensor,
The signal corresponds to the surface of the gear, and the signal is evaluated by the engine controller to calculate the rotational speed;
In the method for calculating the number of revolutions in accordance with the reality, the number of revolutions is calculated from the time interval of the side edge of the set signal.
For the rotation speed larger than the set threshold, the rotation speed is calculated by the first method,
With respect to the rotational speed falls below the threshold value or the second threshold value, out calculate the rotation speed by the second method,
A method for calculating an actual rotational speed, characterized in that an evaluation when the rotational speed is below a set threshold is performed at a shorter time interval than an evaluation when the rotational speed is above a set threshold . 2. The method for calculating an engine speed according to reality according to claim 1, wherein if the engine speed is smaller than a threshold value, the engine speed is calculated from a side edge of each signal or from a side edge of a set signal. If the rotation speed is greater than the threshold value, evaluate the synchronization signal pulse,
The sync pulse are separated from each other by a crankshaft angle which is set,
The method for calculating the engine speed according to the actual claim according to claim 1 or 2, wherein the crankshaft angle depends on the number of cylinders of the engine. The control device has a drum memory having a set number of storage spaces,
Write the time value of each signal side edge to the drum memory,
The reality according to any one of claims 1 to 3, wherein a difference in time value between set signal side edges from the drum memory is evaluated in order to calculate an actual rotation number at a small rotation number. The calculation method of the engine speed according to. The method for calculating the engine speed according to any one of claims 1 to 4, wherein each of the last two entries in the memory is evaluated for calculating the speed. 6. The method for calculating the engine speed according to any one of claims 1 to 5, wherein the time between the last entry in the memory and the nth previous entry is evaluated for calculating the speed. . The method for calculating the engine speed according to the reality according to any one of claims 1 to 6, wherein a calculation time point of the speed is determined by a set determination criterion. Correct and / or compensate when calculating the rotational speed,
The method for calculating the engine speed in accordance with the reality according to any one of claims 1 to 7, wherein disparity due to the system in the course of the speed is taken into consideration. The method for calculating the engine speed according to the reality according to any one of claims 1 to 8, wherein the engine speed is calculated in a set time pattern when the engine speed is small. 10. The value formed for a realistic engine speed (nmot) is additionally averaged and / or filtered and / or limited in a maximum change. A method of calculating the engine speed in line with reality. 11. The method for calculating the engine speed according to any one of claims 1 to 10, wherein the engine speeds calculated by different methods are correlated with each other and are validated using a variable tolerance. In a device for calculating an engine speed that is realistic and includes a calculation device,
12. A device for calculating a realistic engine speed, characterized in that at least one of the methods according to claim 1 is executed in the device.

Images