Camshaft sensor processing method

By introducing qualification determination steps in the camshaft sensor, selecting qualified indexes and optimizing processing task allocation, the problem of computing device congestion caused by tooth edge signal reproducibility is solved, processing efficiency and accuracy are improved, and engine control is ensured. Dynamic and responsive.

CN114190099BActive Publication Date: 2024-11-26VTESCO TECH GMBH
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202080056236.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Priority Date
2019-08-09
Filing Date
2020-07-16
Publication Date
2024-11-26
Estimated Expiration
2040-07-16

AI Technical Summary

Technical Problem

The teeth edge signals in the camshaft sensors exhibit variable reproducibility, resulting in the computing device that may be congested at high engine speeds and cannot process all incoming teeth edges within the allocated time, resulting in processing errors and control biases. shift.

Method used

By introducing the qualification determination step in the camshaft sensor, qualified indexes are selected and retained, and only the second process is performed on these indexes, reducing the number of processing and avoiding congestion in the computing device. Qualification determination may be based on frequency, time, angular distance, or a predefined index list, where the isometric index is judged to be eligible and processing tasks are assigned in a multiprocessor environment to optimize processing efficiency.

Benefits of technology

It effectively reduces the load of the processing device, avoids congestion in the processing device, ensures regular sampling and dynamics of the signal, improves the responsiveness and accuracy of processing execution, and reduces the risk of processing delay at high-speed engine speed.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114190099B_ABST
    Figure CN114190099B_ABST
Patent Text Reader

Abstract

A processing method for a camshaft sensor (1), the camshaft sensor (1) comprising a gear (2) rotating integrally with the camshaft, the gear (2) rotating in front of a fixed sensing element (3), the sensing element (3) being capable of emitting a signal indicating a tooth profile of the gear (2), the method comprising the following steps: receiving a tooth edge, identifying a corresponding index (I1-I8), performing at least one first processing for consuming each index (I1-I8), the method further comprising the following steps: performing a qualification determination on the index (I1-I8), and if an index (I1-I8) is determined to be qualified, performing at least one second processing for consuming the index (I1-I8).
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to processing a camshaft sensor. Background Art

[0002] As is known, a camshaft sensor comprises a gear comprising a small number of teeth, which are most often irregularly and / or irregularly spaced around the circumference. The gear rotates integrally with the camshaft. A sensing element, fixed relative to the engine and capable of detecting a tooth edge, is arranged opposite the gear. The sensing element is capable of emitting a signal indicative of the gear tooth profile, since the signal comprises identical tooth edges. Each tooth edge is characterized by its index.

[0003] The computing device processes the signal. The problem is that the signal containing the tooth edges exhibits a variable reproducibility. In addition, when the engine speed reaches higher values, the computing device may become jammed and the computing device may no longer be able to process all incoming tooth edges within the allotted time. Thus, a new edge may be received before the previous one has been processed.

[0004] The incoming tooth edges are processed in particular to determine the index of each edge, i.e. to identify each edge. This index, which is associated with the engine angular position, is used by the algorithm for controlling the variable distribution. Furthermore, not processing the tooth edges has detrimental consequences, since the perception of the algorithm for controlling the variable distribution will be offset by the same amount. This then causes all processing based on the angular position to be wrong. Summary of the invention

[0005] There is an urgent need for a treatment method that does not have the drawbacks, disadvantages and obstacles of the prior art.

[0006] In order to solve one or more of the above-mentioned shortcomings, the present invention proposes a processing method for a camshaft sensor, wherein the camshaft sensor includes a gear that rotates integrally with the camshaft, and the gear rotates in front of a fixed sensing element, and the sensing element is capable of emitting a signal indicating the gear tooth profile, and each tooth edge of the gear is represented by its index, and the index is the serial number of the tooth edge on the gear of the camshaft sensor. The method includes the following steps: receiving the tooth edge, identifying the corresponding index, and performing at least one first processing that consumes each index. The method also includes the following steps: performing qualification judgment on the index, and the qualification judgment is a selection step, selecting whether to retain an index or not in a set of indexes, and the retained index is then called a qualified index, and if an index is determined to be qualified, performing at least one second processing that consumes the index.

[0007] Specific features or embodiments that may be used alone or in combination are:

[0008] - A qualification step qualifies the indexes according to the following criteria: frequency criterion, 1 index out of n indexes is qualified; minimum time elapsed since the last qualified index; angular distance relative to the last qualified index; or according to a predefined index list,

[0009] - the qualification determination step determines the isometric index as being qualified,

[0010] -The number n is an increasing function of the engine speed,

[0011] - if the method is implemented in a multi-processor or multi-core environment, the identification, eligibility determination and execution of at least one first process are performed on a first processor or core, preferably by a parent task, and the execution of at least one second process is performed on another processor or core, preferably by a child task of the parent task, without triggering child tasks for indexes that are not determined to be eligible,

[0012] - the method further comprises the steps of storing the index a first time so as to be specific to said at least one first process and storing the index at least a second time so as to be specific to said at least one second process,

[0013] - said at least one first process comprises determining the engine speed,

[0014] - said at least one second process comprises the control of a variable allocation. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] The invention will be better understood by reading the following description, which is given by way of example only and with reference to the accompanying drawings, in which:

[0016] Figure 1 The camshaft sensor and its processing device are shown.

[0017] Figure 2 shows a signal S and its processing according to the prior art,

[0018] Figure 3 shows a signal S and its processing according to the invention,

[0019] FIG4a shows a flow chart illustrating the part of the method executed on the first core,

[0020] FIG. 4 b shows a flow chart illustrating the part of the method executed on the second core, the two flow charts being connected by a dashed arrow. DETAILED DESCRIPTION

[0021] refer to Figure 1The camshaft sensor 1 comprises a gear 2, which comprises a small number of teeth, which are usually irregular. The gear 2 rotates integrally with the camshaft. The gear 2 rotates in front of a sensing element 3 which is fixed relative to the engine. The sensing element 3 is capable of detecting tooth edges and is capable of emitting a signal S indicating the tooth profile of the gear 2, the signal S being capable of indicating the tooth profile because the signal S has the same tooth edges as a function of time. Figure 2 or Figure 3 The Figure 1 An example of such a signal S obtained by the camshaft sensor 1. The processing device 4 receives the signal S and performs processing thereof.

[0022] The method for processing such a signal S originating from a camshaft sensor 1 comprises, in a known manner, the following steps: receiving S1 a tooth edge transmitted via the signal S, identifying S2 an index I1-I8 corresponding to the received tooth edge, performing S3 at least one first processing consuming each index I1-I8. The index I1-I8 is the ordinal number of a tooth edge on the gear 2 of the camshaft sensor 1. The index is absolute and periodic over a revolution of said gear 2.

[0023] Figure 1 The gear 2 of the camshaft is marked with an angle, and the unit is ° (angle). One rotation of the gear 2 of the camshaft is represented as 360°. As a function of time, Figure 2 and Figure 3 The signal S adopts a rotation reference corresponding to the crankshaft in degrees, either CRK degrees or CRK. In addition, the signal S is inverted: the rising edge of gear 2 appears to be falling, and vice versa. As is well known, 1 CRK is two 1°, because the camshaft rotates once when the engine / crankshaft (or "crank" in English, CRK) rotates twice. The first index I1 corresponds to a rising edge at 90° or 180 CRK. The second index I2 corresponds to a falling edge at 105° or 210 CRK. The third index I3 corresponds to a rising edge at 120° or 240 CRK. The fourth index I4 corresponds to a falling edge at 195° or 390 CRK. The fifth index I5 corresponds to a rising edge at 240° or 480 CRK. The sixth index I6 corresponds to a falling edge at 285° or 570° CRK. The seventh index I7 corresponds to a rising edge at 0° or 0 CRK. The eighth index I8 corresponds to the falling edge at 15° or 30° RK.

[0024] The processing method may use all tooth edges. In this case, an index is assigned to each tooth edge. Alternatively, the processing method may only use certain tooth edges. According to an example of this embodiment, the processing method uses only rising edges or only falling edges. In this case, the other type of edge is simply ignored in the identification process and is not assigned an index.

[0025] According to the prior art, all processing consuming at least one index I1 - I8 is systematically performed for each index. This can be considered as one of the causes of congestion and problems of the processing device 4 .

[0026] In order to eliminate the congestion of the processing device 4, the present invention proposes to divide the processing that consumes at least one index I1-I8 into at least two categories. The first category of processing includes a first processing that absolutely must consume and process all indexes. Like the prior art, a first processing belonging to the first category is triggered and executed for each index I1-I8. At least the second category of processing includes a second processing that can consume and process only some of all indexes, which are called qualified indexes. Therefore, advantageously, the second processing belonging to this class or these classes is triggered and executed only for qualified indexes. The qualification of the index is determined in the qualification determination step S4.

[0027] The qualification decision S4 is a selection step, which selects whether to retain an index (then called a qualified index) or not to retain the index in the set of indexes I1-I8. The subset of qualified indexes advantageously has a smaller cardinality relative to the set of indexes I1-I8. By triggering and executing only the processing associated with these qualified indexes, the method according to the present invention reduces the number of processing, thereby making it possible to avoid or at least postpone congestion of the processing device 4.

[0028] The qualification decision S4 can be performed according to any selection mode. In addition, a plurality of qualification decisions can be performed according to criteria, which can be different from each other and each associated with one of the specific second processes.

[0029] For example, a frequency criterion may be used to make the qualification decision. In this case, the qualification decision step retains 1 index out of n indices, where n is any integer at least equal to 2. Thus, when the first index is determined to be qualified, the n-1 indexes immediately following it are ignored by the qualification decision step and cannot be determined to be qualified.

[0030] It is also possible to use a criterion of the minimum time elapsed since the last qualified index. Thus, when the first index is determined to be qualified, one or more indexes immediately following it included in the given time window are ignored by the qualification determination step and cannot be determined to be qualified. Such a criterion is advantageous in that the closer in time the index and the possibly associated processing are, the more likely the processing device 4 is to be congested.

[0031] A criterion of angular distance relative to the previous qualified index can also be used. The tooth profile of the gear 2 is known. In addition, when the index I1-I8 corresponding to the tooth edge is identified, the angular distance between the edge and the edge immediately preceding or following it is known. In this way, when the first index is determined to be qualified, one or more indexes that follow it and have a certain angular distance from the qualified index are ignored by the qualification determination step and cannot be determined to be qualified. The angular distance can be a minimum distance. In this way, when the first index is determined to be qualified, one or more indexes that follow it and are included in a given angular window after the qualified index are ignored by the qualification determination step and cannot be determined to be qualified. The advantage of such a criterion is that the closer the index and the possibly associated processing are in time, the more likely the processing device 4 is to be congested, wherein the temporal proximity can be derived from the angular proximity based on the engine speed.

[0032] It is also possible to use a criterion for arbitrarily selecting certain indices, for example according to a predefined list of indices. Such a list is advantageously periodic. Thus, for the list (I1, I6), only indices I1 and I6 can be judged as eligible, which has the effect of reducing the number of second processes by one third.

[0033] According to a particularly advantageous feature, the criterion for determining the eligibility of the index is equidistance. The qualification step determines that the equidistant index is eligible. In most cases, the gear 2 is shaped to have edges that are angularly equidistant, and more advantageously modulo the number of cylinders of the engine. In this way, Figure 1 The gear 2 shown in FIG. 1 has three rising edges, with respective indices 3, 5 and 7 equally spaced at 240°RK or 120°, which is suitable for a three-cylinder engine. It also has four falling edges, with respective indices 2, 4, 6 and 8 equally spaced at 180°RK or 90°, which is suitable for a four-cylinder engine. By only determining equally spaced indices as eligible, the method reduces the number of second processes to 3 / 8 (3 cylinders) or 1 / 2 (4 cylinders), respectively.

[0034] The advantage of the equidistant indices is that they are regularly spaced in angle and synchronized with the operation of each cylinder of the engine. This ensures a regular sampling of the signal, thus guaranteeing equal dynamics of the result. Therefore, the fact of considering equidistant indices ensures a subsampling with respect to all indices I1-I8, which is advantageously according to a regular pseudo-period. Moreover, in the case of constant engine speed, the times measured between these equidistant indices are equal.

[0035] It is also possible to combine the qualification criteria with each other. In this way, one of the n indices of the verification time and / or angular distance criteria can be determined as qualified. It is also possible to retain only the index of the verification time distance criterion, and from the retained indexes, only one of the n indices can be determined as qualified. Here, those skilled in the art will know how to add a combination of two or more criteria.

[0036] According to another feature, instead of a constant n, a frequency criterion can be applied to a function number n of the engine speed, advantageously an increasing function of the engine speed. Thus, for example, at low speeds, n can be taken equal to 1. When the engine rotates slowly, all indexes are judged to be eligible. At low speeds, there is almost no risk of congestion of the processing device 4. However, a minimum number of indexes is necessary to ensure the dynamics of the results of one or more second processes. At higher speeds, n can be taken equal to 3. When the engine rotates faster, the number of eligible indexes can be reduced. This can reduce the risk of congestion of the processing device 4 without interfering with the dynamics of the results. At high speeds, n can be taken equal to 10. When the engine rotates further faster, the number of eligible indexes can be further reduced. This can reduce the risk of congestion of the processing device 4 without interfering with the dynamics of the results. It can be inferred that such a feature makes it possible to correct the influence of the engine speed by trying to average the pseudo-frequencies of the eligible indexes. The previous example used three engine speed levels, but any number can be used.

[0037] This last feature can obviously be combined with other qualification criteria.

[0038] Thus, according to a particularly advantageous feature, only the equidistant indices can be retained and their eligibility determined based on a frequency criterion, where n is a function of the engine speed. Thus, for example, n can be taken to be equal to 1 at low speed, equal to 2 at medium speed, and equal to 4 at high speed. Thus, for a 4-cylinder engine, all the equidistant indices determined to be eligible are: I2, I4, I6, I8 at low speed, only the indices I2, I6 or I4, I8 at medium speed, and only one of the four indices at high speed. Here, the equidistant indices are advantageous in that the indices selected in this way according to the frequency criterion are still equidistant.

[0039] Currently, the engine control processing and the camshaft sensor processing therein are implemented by a multi-processor processing device 8, which includes multiple processors or includes at least one multi-core processor.

[0040] In this case, identification S2, eligibility determination S4 and execution S3 of at least one process consuming each index are advantageously performed on a first processor or core, preferably by a parent task, and execution S9 of at least one process consuming the qualified index is advantageously performed on another processor or core, preferably by a child task of the parent task.

[0041] Figure 2 and Figure 3 The operation of such a multi-processor or multi-core device is shown. A first row corresponding to a first process executed by a first core and a second row corresponding to a second process are plotted below the signal S and opposite each other.

[0042] Figure 2 It shows the congestion problem that may occur when no qualification decision is made on the indexes and the second process is performed on each index.

[0043] In the nominal case, after receiving edge 4, the corresponding parent task M4 is triggered. The parent task M4 recognizes edge: I4, records the index in the memory shared with the child task, and triggers the execution of the corresponding child task F4. The corresponding child task F4 re-reads the index I4 in the shared memory and consumes it to perform its second processing. For the next receiving edge 5, the corresponding parent task M5 is triggered. The parent task M5 recognizes edge: I5, records the index in the shared memory, and triggers the execution of the corresponding child task F5. The corresponding child task F5 re-reads the index I5 in the shared memory and consumes it to perform its second processing.

[0044] When two consecutive edges become too close in time, due to the angular proximity of the edges on gear 2 and the increase in the rotational speed of the camshaft and gear 2, the following problematic situation may occur. After receiving edge 7, the corresponding parent task M7 is triggered. Parent task M7 recognizes edge: I7, records the index in the memory shared with the subtask, and triggers the execution of the corresponding subtask F7. However, subtask F7 is sufficiently delayed so that the next edge 8 arrives. When this next edge 8 is received, the corresponding parent task M8 is triggered. Parent task M8 recognizes edge: I8, records the index in the shared memory, and triggers the execution of the corresponding subtask F8. This second record is superimposed on the previous index I7 in the shared memory and thus erases the index I7. In addition, subtask F7 re-reads index I8 in the shared memory when it is finally executed, instead of the nominal index I7. This modification will disrupt the correct execution of subtask F7, which performs processing on inconsistent information. Subtask F8 also re-reads index I8 in the shared memory and consumes it to perform its second processing.

[0045] In contrast, Figure 3The use of qualification determination is shown. The use of index qualification determination list: I3, I5 and I7, where they are equidistant. After receiving edge 7, the corresponding parent task M7 is triggered. Parent task M7 identifies edge: I7. Since edge I7 is determined to be qualified, parent task M7 records the index in the shared memory of the subtask and triggers the execution of the corresponding subtask F7. When the next edge 8 is received, the corresponding parent task M8 is triggered. Parent task M8 identifies edge: I8. However, this index I8 is not determined to be qualified, and parent task M8 does not record the index in the shared memory or records it in a different shared memory to protect the recording of the previous index I7 and does not trigger the execution of subtask F8. In addition, subtask F7, even if it is postponed relative to parent task M8, re-reads the index I7 in the shared memory and can correctly perform its second processing. The same is true for other qualified indexes I3 and I5 with corresponding tasks M3, F3, M5, and F5.

[0046] The eligibility determination thus makes it possible to remove unnecessary subtasks and thus makes it possible to reduce the risk of congestion of the second core, thereby limiting the risk of delays of subtasks. More generally, the main advantage of the invention is that it determines that a subset of indices I1-I8 is eligible and reduces the load on the processing device 4 by removing a large number of indices from the second processing. This advantageously allows the execution of processes that do so and therefore do not affect the operation. The possible cores or processors on which these removed processes are executed therefore become more responsive. The waiting delay between parent and child tasks is thus improved. Advantageously, this means that the execution of the process retained because it is determined to be eligible (because it corresponds to the index determined to be eligible) is closer in time to the engine position associated with it.

[0047] According to the first feature seen in the previous example, during identification, the parent task does not store in the shared memory the indexes that were not judged to be eligible, to avoid any risk of re-posting, since such re-posting would erase the previous posting and could cause consistency problems. This is a first mechanism that allows protecting the eligible index posted to the shared memory until it is re-read by the child task that performs the second processing.

[0048] According to another alternative feature, if it is identified that the index is systematically entered / stored in the first shared memory, at least one second shared memory is advantageously provided for storing only the qualified indexes. This forms another mechanism for protecting the qualified indexes. It is identified that the index is stored in the first shared memory for the first time, dedicated to the at least one first process. And in the case of a qualified index, it is identified that the qualified index is stored in the at least one second shared memory at least a second time, dedicated to the at least one second process.

[0049] A single first memory can advantageously be used for the first storage. For the at least one second shared memory, it is advantageous to provide at least one for each second process. This is particularly advantageous if the second processes use different eligibility decisions in order to protect their respective eligible indexes from each other.

[0050] The first processing is a processing that systematically consumes all indices. The first processing necessarily includes identification S2, but also includes qualification determination S4. A symbolic first processing includes determining the engine speed. This determination is usually to redundancy check (redonder) the nominal determination performed by the crankshaft sensor processing. Due to the small number of edges on gear 2, it is preferred to use all edges / indexes. This processing is advantageously performed by the parent task.

[0051] The second process is a process that consumes only eligible indexes. An example of a second process is the control of variable allocation. This process is advantageously performed by a subtask.

[0052] Variable distribution is a known technology that, by varying the angular position of the camshaft relative to the engine / crankshaft, allows to vary several parameters that influence engine control: timing (in English "variable valve timing", better known by its acronym VVT), opening and / or lift duration (in English "variable valve lift") of the intake and exhaust valves. These parameters are driven according to speed, load and acceleration requirements. The benefits of variable distribution are high torque at low speed, high power at high speed, greater efficiency (allowing the engine to run in the Atkinson cycle and reducing pumping losses) and less pollution.

[0053] The VVT ​​changes with the speed. It phase-shifts the camshaft relative to the crankshaft. This phase shift is controlled by the engine control. The slave control of the VVT ​​is an example of a second process, advantageously a second process executed by a subtask of the processing device 4 consuming a qualified index.

[0054] Fig. 4a and Fig. 4b show a flow chart of an embodiment of a processing method according to the present invention. Fig. 4a shows a step preferably performed by a parent task / on a first core. The first step S1 receives a signal S and detects an edge. The second step S2 identifies the index corresponding to the edge. The third step S3 performs one or more first processes, i.e., processes that consume all indexes to perform. The fourth step S4 checks whether the index identified by the previous step S2 is determined to be eligible. If it is not determined to be eligible, the method branches to end S7 and terminates. If the index is determined to be eligible, the method branches to step S5, and step S5 stores the eligible index and optionally the data useful for (one or more) second processes in at least one shared memory. The method then continues to perform (one or more) second processes. If the method is performed in a multi-processor or multi-core environment, this is performed by transferring control to a subtask, which is performed by step S6, and step S6 activates the subtask executed on another core.

[0055] Figure 4b shows the steps of executing the subtask on another core. Step S8 re-reads the eligible index and data stored in step S5 from the shared memory. Step S9 performs one or more second processes.

[0056] The invention has been illustrated and described in detail in the drawings and the foregoing description. This should be considered illustrative and given in the name of example and not to limit the invention to this description. Numerous implementation variations are possible.

[0057] Reference numerals list

[0058] 1: Camshaft sensor,

[0059] 2: Gear,

[0060] 3: Sensing element,

[0061] 4: Processing device,

[0062] S: signal from sensing element 3,

[0063] I1-I8: index,

[0064] F3, F4, F5, F7, F8: subtasks,

[0065] M3, M4, M5, M7, M8: parent tasks,

[0066] S1-S9: Steps of the method.

Claims

1. A processing method for a camshaft sensor (1), the camshaft sensor (1) comprising a gear (2) rotating integrally with the camshaft, the gear (2) rotating in front of a fixed sensing element (3), the sensing element (3) being capable of emitting a signal (S) indicative of a tooth profile of the gear (2), each tooth edge of the gear being characterized by its index (I1-I8), the index (I1-I8) being the serial number of the tooth edge on the gear (2) of the camshaft sensor (1), the method comprising the following steps: Receive a tooth edge (S1), identify (S2) the corresponding index (I1-I8), and execute (S3) at least one first process that consumes each index (I1-I8), characterized in that the method also includes the following steps: perform qualification judgment (S4) on the index (I1-I8), the qualification judgment (S4) is a selection step, and selects whether to retain an index or not in the set of indexes (I1-I8), and the retained index is then called a qualified index, and if an index (I1-I8) is judged to be qualified, execute (S9) at least one second process that consumes the index (I1-I8).

2. The method according to claim 1, wherein: The qualification determination (S4) step determines the qualifications of the indexes (I1-I8) according to the following criteria: frequency criterion, determining that 1 index among n indexes is qualified; the minimum time elapsed since the previous qualified index; the angular distance relative to the previous qualified index; or according to a predefined index list.

3. The method according to claim 2, wherein: The qualification determination (S4) step determines that the equidistant index is qualified.

4. The method according to any one of claims 2 or 3, wherein: The number n is an increasing function of the engine speed.

5. The method according to any one of claims 1 to 3, implemented in a multi-processor or multi-core environment, wherein: The identification (S2), qualification determination (S4), and execution of at least one first process (S3) are performed on a first processor or core, and the execution of at least one second process (S9) is performed on another processor or core.

6. The method according to claim 5, wherein: Identification (S2), qualification determination (S4) and execution of at least one first process (S3) are performed by a parent task, and execution of at least one second process (S9) is performed by a child task of the parent task, without triggering the child task for an index that is not determined to be qualified.

7. The method according to any one of claims 1 to 3, further comprising the steps of: The indexes (I1-I8) are stored (S5) a first time to be specific to the at least one first process, and the indexes are stored separately at least a second time to be specific to the at least one second process.

8. The method according to any one of claims 1 to 3, wherein: The at least one first process includes determining an engine speed.

9. The method according to any one of claims 1 to 3, wherein: The at least one second process comprises control of a variable allocation.

Citation Information

Patent Citations

  • Geberrad sowie Vorrichtung und Verfahren zum Ermitteln der Position einer Brennkraftmaschine

    CN104895673A

  • Hydraulic VVT angle dynamic measuring method of vehicle engine

    CN105888842A

  • Method and apparatus for controlling a vehicle engine

    GB9223918D0

  • Method for synchronizing an internal combustion engine based on the angular position of a rotating part

    US20020157649A1

  • Method and Control Unit for Controlling an Internal Combustion Engine

    US20130206108A1