JP4684066B2 - Ignition timing control device for internal combustion engine - Google Patents

Description

  The present invention relates to an ignition timing control device for an internal combustion engine that controls the ignition timing of the internal combustion engine in accordance with the state of occurrence of knocking that occurs in the internal combustion engine.

  In general, knocking (hereinafter referred to as “knock”) control of an internal combustion engine is knocked when an electrical signal from a knock sensor that detects engine vibration exceeds a predetermined level (knock determination level). The ignition timing is retarded, and when the knock is not detected for a predetermined period, the ignition timing is advanced so that the ignition timing is always controlled near the knock limit, and the fuel consumption and output characteristics of the engine are controlled. It is the one that brings out the maximum.

  In this case, conventionally, a knock sensor for detecting knock vibration is attached to the cylinder block of the internal combustion engine, and a knock frequency component is extracted from the output signal of the knock sensor for each combustion (one ignition) by a band-pass filter. The occurrence of knock is detected by comparing the peak value of each knock frequency component with the threshold for knock determination. If a knock is detected, the ignition timing is retarded by a predetermined amount, and if knock is not detected, it is predetermined. Ignition timing control was performed by attenuating (advancing) the amount of retardation every time.

  Since the knock detection result for each combustion is used in this way, if it is erroneously detected that knock has not occurred due to a detection error or the like, the ignition timing is erroneously retarded and the engine output is Cannot be pulled out sufficiently. On the other hand, if a knock has occurred but is not detected, the ignition timing may be mis-advanced, resulting in frequent knocks, which may lead to engine damage.

  For this reason, the present applicant uses a knock occurrence state determination method that is separate from the above-described method using knock feedback control that performs knock determination by comparing the peak value of the knock frequency component for each combustion with the threshold for knock determination. In addition, it is proposed that each method is operated independently, and that the threshold value used for knock detection for each combustion (ignition) is corrected based on the knock generation state determination result.

  As described above, in the above-described control method in which the two methods are operated independently, even if an erroneous retardation occurs due to an initial matching error in the knock determination threshold value for each combustion, the other knock generation state determination method is used. When it is determined that no knock has occurred, the knock determination threshold can be appropriately corrected, so that the robustness is improved as compared with the case where the conventional knock occurrence state determination method is not used. .

  However, when trying to control to the state where a minute knock is generated, the conventional method of operating the knock generation state determination means based on the conventional knock detection result does not allow sufficient delay angle correction because the frequency of knock occurrence is low. In order to control to a state where a minute knock is generated, it is necessary to increase a feedback gain (retard amount) in knock feedback control for each combustion. However, this leads to an increase in the ignition timing fluctuation range. This ignition timing fluctuation is undesirable because it leads to fluctuations in output torque and deterioration in emissions.

  The present invention has been made in view of the above problems, and its object is to retard the vicinity of the ignition timing at which a minute knock is generated, and to reduce the fluctuation range of the ignition timing by knock feedback control. An ignition timing control device for an internal combustion engine is provided.

The present invention provides an ignition timing control device for an internal combustion engine according to claims, as means for solving the problems.
The ignition timing control apparatus for an internal combustion engine according to claim 1 is a knock sensor that outputs a signal having a vibration waveform corresponding to a knock state of the internal combustion engine, and a knock signal generated by the knock signal output by the knock sensor. Based on the statistical characteristics of the knock sensor output for a plurality of ignitions obtained by this knock sensor, and feedback control means for calculating occurrence and calculating the ignition timing retard amount or advance amount (eaknk) based on the determination result Ignition timing learning control means for determining a knock occurrence state and calculating a learning amount (eaknk) obtained by retarding or advancing the ignition timing based on the determination result, the feedback control means and the ignition timing The final ignition timing of the internal combustion engine is calculated based on the learning control means. Thus, by learning the ignition timing based on the knock generation state determination that does not depend on the knock detection for each ignition, it is possible to correct the delay to the vicinity of the ignition timing at which the minute knock generation state occurs, and the ignition timing by the knock feedback control The fluctuation range can be reduced.

Hereinafter, an ignition timing control device for an internal combustion engine according to an embodiment of the present invention will be described with reference to the drawings.
Based on FIG. 1, a schematic configuration of an entire engine control system including an ignition timing control device according to the present invention will be described. An air cleaner 13 is provided at the most upstream portion of the intake pipe 12 of the engine 11 that is an internal combustion engine, and an air flow meter 14 for detecting the intake air amount is provided downstream of the air cleaner 13. On the downstream side of the air flow meter 14, a throttle valve 15 whose opening is adjusted by the motor 10 and a throttle opening sensor 16 for detecting the throttle opening are provided.

  Further, a surge tank 17 is provided on the downstream side of the throttle valve 15, and an intake manifold 19 for introducing air into each cylinder of the engine 11 is provided in the surge tank 17, and intake air of the intake manifold 19 of each cylinder is provided. Fuel injection valves 20 for injecting fuel are attached in the vicinity of the ports. A spark plug 21 is attached to each cylinder of the engine 11 for each cylinder, and the air-fuel mixture in the cylinder is ignited by spark discharge of each spark plug 21.

  On the other hand, the exhaust pipe 22 of the engine 11 is provided with a catalyst 23 such as a three-way catalyst that purifies CO, HC, NOx, etc. in the exhaust gas, and detects the air-fuel ratio of the exhaust gas upstream of the catalyst 23. An air-fuel ratio sensor 24 is provided. In addition, the cylinder block of the engine 11 includes a coolant temperature sensor 25 that detects a coolant temperature, a knock sensor 28 that detects knock vibration, and a crank that outputs a pulse signal each time the crankshaft of the engine 11 rotates a predetermined crank. An angle sensor 26 is attached. Based on the output signal of the crank angle sensor 26, the crank angle and the engine speed are detected. Instead of the knock sensor 28, an in-cylinder pressure sensor (not shown) may be used.

  Outputs of these various sensors are input to an engine control unit 27 (hereinafter referred to as “ECU 27”). The ECU 27 is mainly composed of a microcomputer, and executes various engine control programs stored in a built-in ROM (storage medium), thereby allowing the fuel injection amount of the fuel injection valve 20 and the ignition timing of the spark plug 21 to be executed. To control. The RAM built in the ECU 27 is used for storing data when executing each program and various predetermined values.

  The ECU 27 includes an ignition timing learning control means for determining the presence or absence of a knock generation state by executing a program for determining a knock generation state, which will be described later, and controlling the ignition timing based on the determination result. Further, the ECU 27 compares the peak value of the knock frequency component of the sensor output for each combustion, which will be described later, with the knock determination threshold value to determine the presence or absence of knock for each combustion, and controls the ignition timing based on the determination result. Includes feedback control means. When it is determined that there is a knock, the ignition timing is retarded to suppress the knock, and when it is determined that there is no knock, the ignition timing is advanced. As a result, the ignition timing is advanced within the range of the knocking sound that is permissible for hearing, so that the engine output and fuel consumption are improved.

FIG. 2 is a flowchart showing the overall flow of ignition timing control by the ignition timing control device for an internal combustion engine of the present invention, FIG. 3 is a diagram for explaining the final ignition timing in the ignition timing control, and FIG. It is a timetable of the whole timing control.
In the internal combustion engine ignition timing control apparatus of the present invention, as shown in FIG. 2, feedback control by the feedback control means (step S1) and ignition timing learning control by the ignition timing learning control means (step S2) are performed in parallel. Thus, the final ignition timing (step S3) is determined. That is, as shown in FIG. 3, the final ignition timing (eaop) is determined from the base ignition timing (eabse) by the retard amount or the advance amount calculated by the feedback control and the retard angle correction or the advance angle correction by the ignition timing learning control. Is obtained by subtracting the learned amount (eagknk). Note that feedback control and ignition timing learning control will be described later in detail.

  A time table of the entire ignition timing of the present invention thus performed is shown in FIG. In FIG. 4, the number of combustions (ignition), knock detection for each combustion (ignition), knock feedback control retardation amount (eaknk), knock state determination, ignition timing learning amount (eakg), final ignition timing ( eaop). That is, it is determined that knocking has occurred in the second, fourth, and fifth combustion, respectively, and the retard processing is performed by knock feedback control. The portion that is rising to the right indicates that the advance angle process is being performed gradually because it is determined that no knock has occurred. When the data of combustion up to N times is accumulated, it is determined whether or not the combustion state in this period is a knock state. In FIG. 4, the first N knock state determinations are determined as knock occurrences, and the ignition timing is offset by the learning amount (eakg). In the first N combustions, six knocks were detected and the ignition timing was retarded. However, in the next N combustions, the same delay was detected only by detecting two knocks. Since the angular amount can be realized, fluctuations in the ignition timing can be suppressed. Therefore, the engine torque can be obtained stably.

  FIG. 5 is a flowchart showing the procedure of knock feedback control. First, knock is determined from the output signal detected by the knock sensor 28 in step S100. This knock determination is performed based on the knock detection determination flowchart shown in FIG. That is, as shown in FIG. 6, the peak value P of the knock frequency component obtained from the output signal of the knock sensor 28 is first detected in step S101. Next, in step S102, the peak value P is compared with a knock determination threshold value (knkjg). When the peak value P is larger than the knock determination threshold value (YES), it is determined that there is a knock in step S103, and when the peak value P is smaller than the knock determination threshold value (NO), it is determined that there is no knock in step S104.

  Next, returning to the flowchart of FIG. 5, the knock detection determination (S100) is performed as described above, and if it is determined that the knock has occurred in step S110 (YES), the process proceeds to step S111, where the advance side is greater than the maximum retard amount. It is determined whether or not. If it is determined that the angle is on the advance side (YES), the process proceeds to step S112, and a retard process of the retard amount (eank) added by the predetermined amount Ar is performed. This predetermined amount Ar is determined according to the magnitude of the peak value P, for example. Next, in step S113, it is determined whether a predetermined time has elapsed after the retard processing. If it is determined in step S110 that there is no occurrence of knock (NO) and if it is determined in step S111 that it is not on the advance side of the maximum retardation amount (NO), the process proceeds to step S113.

  If it is determined in step S113 that the predetermined time has elapsed (YES), the process proceeds to step S114, and it is determined whether or not it is on the retard side from the base ignition timing (eabse). If it is determined that the angle is on the retard side (YES), the process proceeds to step S115, and the advance processing of the advance amount (eaknk) drawn by the predetermined amount Aa is performed. After the advance angle processing, the process proceeds to step S116, the timer is reset, and time measurement starts again. If it is determined in step S113 that the predetermined time has not elapsed (NO), the flowchart ends. If it is determined in step S114 that the timing is not retarded from the base ignition timing (NO), the process proceeds to step S116, and the timer is reset. In this way, the knock feedback control of the sawtooth ignition timing as shown in FIG. 4 is performed.

  FIG. 7 is a flowchart showing the procedure of ignition timing learning control. In the ignition timing learning control, N times of combustion data is accumulated, and based on this, first, the knock occurrence state is determined in step S200. One example of this knock occurrence state determination technique is shown in FIG.

  Next, the knock occurrence state determination method will be described with reference to FIG. In this knock occurrence state determination method S200, after performing the process of counting the combustion number N by the combustion number counter and recording this (step S201), the vibration intensity M recorded in advance in steps S202 and S203. The average value Vav and the variance Xdiv are updated. Specifically, these update operations are performed based on the following formulas (1) and (2) using smoothing constants α and β (0 <α, β <1). Further, in step S204, the standard deviation σ is calculated based on the following equation (3).

  Vav ← α × vibration intensity + (1−α) × Vav (1)

  Xdiv ← β × (vibration intensity−Vav) ^ 2 + (1−β) × Xdiv (2)

  σ ← (Xdiv) ^ 0.5 (3)

  By the way, FIG. 9 is a figure for demonstrating the area | region used by the knock generation state determination method. In the typical vibration intensity distribution shown in FIG. 9, in addition to the average value Vav and the standard deviation σ, three regions ZA, a region ZC, and a region ZD are shown. As illustrated, the region ZA is a region larger than “Vav + σ”, and the region ZD is a region smaller than “Vav−σ”. Further, the region ZC is a region located between “Vav−σ” and “Vav”.

  The reason why such area division is performed will be described with reference to FIG. FIG. 10A is a diagram for explaining a knock occurrence state determination method. FIG. 10A shows a state in which the knock intensity increases from the vibration intensity distribution D1 to the vibration intensity distribution D4. As can be seen from FIG. 10A, when the knock intensity increases from the vibration intensity distribution D1 to the vibration intensity distribution D4, the vibration intensity distribution extends in the right direction, that is, the direction in which the peak increases. As a result, the average value Vav also shifts in the direction of increasing, and the standard deviation σ increases. As a result, when the knock intensity increases, for example, as shown in the vibration intensity distribution D3, most of the vibration intensity distribution is included in the region ZC.

  Therefore, in the knock occurrence state determination method of the present invention, the ratio ZD / ZC between the region ZC and the region ZD is calculated for each of the vibration intensity distribution D1 from the vibration intensity distribution D1, and the knock is generated through the ratio ZD / ZC. The presence or absence of a state is determined. In this regard, in this method 200, instead of the area ratio, the number of times the detection level L is exceeded in each of the areas ZA, ZC, ZD, that is, the ratio Nd of the counter values Na, Nc, Nd and the counter values Nd, Nc. / Nc is used.

  FIG. 10B is a diagram showing the relationship between the knocked state, the counter value ratio Nd / Nc, and the counter value Na. As shown in FIG. 10 (b), the ratio Nd / Nc is relatively large in a state where no knock has occurred, and gradually decreases as knock occurs and its strength increases. As shown in the figure, when the knock strength further increases and the so-called “small knock” state is reached, the ratio Nd / Nc takes a constant value (generally zero), and the knock strength increases more than “small knock”. Even if “divergence knock” occurs, the ratio Nd / Nc does not change. In other words, the ratio Nd / Nc is effective for determining the presence or absence of a knock occurrence state in a region where the knock intensity is smaller than “small knock”.

  On the other hand, when attention is paid to the area ZA in FIG. 10A, the vibration intensity distribution in the area ZA decreases as the knock intensity increases. Therefore, as shown in FIG. 10B, the counter value Na in the region ZA also decreases as the knock strength increases. This counter value takes a constant value in the “small knock” state. In the vibration intensity distribution D4 at the time of “divergence knock”, the vibration intensity distribution is distorted and the standard deviation σ is increased, but the distortion of the vibration intensity distribution is more than that, and as shown in FIG. The counter value Na increases again. Therefore, the counter value Na can determine whether or not the “divergence knock” occurrence state exists.

  With reference to the above description, the knock occurrence state determination in the knock occurrence state determination method S200 will be described with reference to FIG. 8 again. In step S205 to step S207 in FIG. 8, the case division into the region ZA, the region ZC, and the region ZD shown in FIG. 10A is performed. That is, in step S205, the vibration intensity M in the region ZD is selected through determination of whether or not the vibration intensity M is smaller than “Vav−σ”, and in step S206, the vibration intensity M is “Vav−σ” or more. The vibration intensity M in the region ZC is selected through determination of whether or not it is smaller than “Vav”. Similarly, in step S207, the vibration intensity M in the region ZA is selected through determination whether the vibration intensity M is larger than “Vav + σ”. Next, in each of step S208 to step S210, processing for recording the counter values Nd, Nc, Na relating to the areas ZD, ZC, ZA is performed.

  Next, in step S211, it is determined whether or not the counter value of the combustion number counter has exceeded a predetermined number. Then, when the predetermined number of times is exceeded, the knock occurrence state is determined by the knock occurrence state determination method.

  First, in step S212, the ratio Nd / Nc is compared with a predetermined value related to the ratio Nd / Nc. If it is determined that the ratio Nd / Nc is not smaller than the predetermined value, it is determined in step S214 that there is no knock occurrence state. (See FIG. 10 (b)). On the other hand, if it is determined that the ratio Nd / Nc is smaller than the predetermined value, it is determined that a knock has occurred, and the process proceeds to step S213.

In step S213, it is determined whether the knock occurrence state is “small knock” or “divergence knock”. Counter value Na is the absence small fence than the predetermined value relates to the counter value Na is knock in step S215 is determined to be "small knock". On the other hand, the counter value Na is when small differences than the predetermined value, knock in step S216 is determined to be "divergent knock". Thereafter, in step S217, the counter values N, Na, Nc, and Nd are reset, and the processing of the knock occurrence state determination method S200 ends. As described above, in the knock occurrence state determination method S200, it is possible to distinguish between the no knock state, the small knock state, and the divergent knock state by using the ratio Nd / Nc and the counter value Na.

  Returning to the flowchart of FIG. 7 again, if the knock occurrence state is determined to have occurred in step S220 by the knock occurrence state determination method S200 as described above (YES), the process proceeds to step S221 and the maximum retardation amount is obtained. It is determined whether or not it is on the more advanced side. If it is determined that the angle is on the advance side (YES), the process proceeds to step S222, and the retardation is corrected to a learning value (eagknk) obtained by adding a predetermined amount B to the base learning value (eakgknk). That is, as shown in FIG. 4, the ignition timing is offset by the learning amount (eakg) in the Nth combustion. If it is determined in step S221 that there is no advance side of the maximum retard amount (NO), the flowchart is ended as it is.

  When it is determined in step S220 that the knock occurrence state is no knock occurrence (NO), the process proceeds to step S223, and it is determined whether or not it is on the retard side with respect to the base ignition timing (eabse). When it is on the retard side (YES), the process proceeds to step S224, and the advance angle is corrected to the learning value (Eagknk) obtained by subtracting the predetermined amount B from the base learning value (eakgknk). If it is determined in step S223 that the timing is not retarded from the base ignition timing (NO), the flowchart is terminated as it is.

  As described above, in the present invention, the knock detection logic and the knock occurrence state determination logic used for the knock feedback control are operated independently, and the knock occurrence state determination result is used for the ignition timing learning control. As described above, by learning the ignition timing based on the knock generation state determination that does not depend on the knock detection for each ignition, it is possible to correct the delay to the vicinity of the ignition timing at which the minute knock generation state occurs, and the ignition timing by knock feedback control. The fluctuation range can be reduced.

It is a schematic block diagram of the whole engine control system containing the ignition timing control apparatus based on this invention. It is a flowchart explaining the flow of the whole ignition timing control which the ignition timing control apparatus of the internal combustion engine of embodiment of this invention performs. It is a figure explaining the last ignition timing of the ignition timing control in this invention. It is a time table of the whole ignition timing control of this invention. It is a flowchart which shows the procedure of knock feedback control. It is a flowchart which shows the procedure of knock detection determination. It is a flowchart explaining the flow of ignition timing learning control. It is a flowchart which shows the procedure of the knock generation state determination method. FIG. 8B shows a procedure continued from FIG. 8A. It is a figure for demonstrating the area | region used by the knock generation state determination method. (A) It is a figure for demonstrating the knock generation state determination method. (B) It is a figure which shows the relationship between knock state, ratio Nd / Nc of counter value, and counter value Na.

Explanation of symbols

DESCRIPTION OF SYMBOLS 11 Engine 20 Fuel injection valve 21 Spark plug 24 Air-fuel ratio sensor 25 Coolant temperature sensor 26 Crank angle sensor 27 Engine control unit (ECU)
28 Knock sensor

Claims (1)

In an ignition timing control device for an internal combustion engine that detects knocking occurring in the internal combustion engine and controls the ignition timing of the internal combustion engine based on the detection result,
A knock sensor that outputs a vibration waveform signal corresponding to the knocking state of the internal combustion engine;
Feedback control means for determining the occurrence of knocking based on a knock signal for each ignition output by the knock sensor, and calculating an ignition timing retard amount or an advance angle amount (eaknk) based on the determination result;
A knocking occurrence state is determined based on a statistical property of knock sensor output for a plurality of ignitions obtained by the knock sensor, and a learning amount (eagknk) in which the ignition timing is retarded or advanced is corrected based on the determination result. Ignition timing learning control means for calculating ,
The final ignition timing (eaop) of the internal combustion engine is calculated from the base ignition timing (eabse) by the retard amount or the advance amount calculated by the feedback control means and the ignition timing learning control means. An ignition timing control device for an internal combustion engine, which is calculated by subtracting the learned amount .

Images