JPS61157765A - Ignition timing control system for internal-combustion engine - Google Patents

Abstract

PURPOSE:To reduce a knocking occurrence rate, by a method wherein, when given learning conditions are statisfied in the same operation region, after the whole of an ignition time control region is rewritten into a current correction value, partial correction is effected. CONSTITUTION:During operation of a vehicle, it is decided according to outputs from a suction air pressure sensor 1 and a sensor 4 for the number of revolutions of an engine whether a correction value is within a control range A of a ROM 17 in an MPU 18. When decision is YES, the number of correcting times in said operation region is checked. When the number of correcting times is increased to a specified value, the correction value of an operation region at a current point of time is rewritten in relation to the whole of the range A to reset the number of correcting times. Thereafter, control is performed so that the value of an ignition time is corrected as learning of each operation region is effected. This enables focusing of the value of an ignition time in a short time, and permits reduction of a knocking occurrence rate.

Description

[Detailed description of the invention] [Industrial application field]

The present invention detects intake air pressure and engine speed and sets the ignition timing value at that time, and when knocking is detected,
The present invention relates to an ignition R period sub-control method for an internal combustion engine that corrects the value of the ignition II period.

[Prior art]

Internal combustion V1 engine allows knocking so that it can always operate at maximum torque under any operating conditions. Efforts have been made to control the ignition timing at the maximum advance angle that can be kept within the range. Therefore, in the method of controlling the ignition timing 1III by electronic control, a map with intake air pressure and engine speed as parameters is created in the ROM, and the basic ignition timing that becomes the model and model is written there. If knocking occurs during operation at that value, a method has been adopted that calculates a correction value and partially corrects the ignition timing value (Mochima No. 55-91765). Public bulletin). The problem with this method is that after the start of operation, the ignition timing control
A study is performed on the entire range, and it takes a considerable amount of time to obtain the optimum ignition value for each operating range.

[Problems to be solved by the invention]

The present invention was made based on the above 5a situation, and initially under the same conditions, for example, if the number of corrections reaches a predetermined chain, the learned value at this time is set to the ignition timing system t.
By applying it to the correction values of all 11 regions to approximate the optimal ignition timing value at an early stage, progressing to a certain stage of ignition timing control in a short period of time, and then entering partial correction, it is possible to Internal combustion that aims to converge the ignition timing value, lowers the non-king occurrence rate, and improves the driving feeling! ! ! This is an attempt to provide a new ignition timing control system. [Means for Solving the Problems] For this purpose, the present invention detects the intake air pressure and engine speed, sets the ignition timing value at that time, and corrects the ignition timing value when knocking is detected. In the ignition timing control region, the map is equipped with a map that uses intake air pressure and engine speed as parameters, and when the above-mentioned ignition timing value is learned and predetermined learning conditions are satisfied for the same operating region. The present invention is characterized in that the entire ignition timing control region is rewritten to the correction value at that time, and then the ignition timing value is corrected while learning for each operating region.

【Example】

Hereinafter, one embodiment of the present invention will be described in detail with reference to the drawings. In FIG. 1, reference numeral 1 is a sensor that detects intake air pressure (or air amount) such as suction pipe negative pressure, and the sensor output is input to the A/DIl converter 3 via a buffer 2, where it is Digital signal is converted. Also, code 4 is
This is a sensor such as a crank sensor that detects the engine rotation speed, and the output of the sensor is input to the interrupt processing circuit 6 via the buffer 5. And the above sensor output is
input to microprocessor 18; On the other hand, when knocking occurs, a knock signal is captured by the knock sensor 7 and introduced into the microprocessor 18. At this time, the filter 8 first detects noise caused by steady vibrations such as vibrations of the valve train while the engine is running. It is necessary to cut off. Also, due to differences in engine speed when non-king occurs, the knock signal level changes and the noise level also changes, so the output of the filter 8 is divided into two, one input to the amplifier 9, and the other input to the M flow. - It is averaged via the integrating circuit 10, amplified by the amplifier 11 to adjust the level, and compared by the comparator 12 to determine and extract the knock signal. The internal configuration of the microprocessor 18 is as known in the art, consisting of an input port 13, an output port 14. CPIJ15
゜RAM16. The ROM 17 is connected by a pass line, and the input port 13 is connected to the microprocessor 18.
A matched sensor signal accepted by the ignition device 21 is inputted, and a control signal is outputted from the output port 14 to an output circuit 19, and a drive signal is outputted from the output circuit 19 to the ignition device 21. As shown in FIG. 2, the ROM 17 is provided with a map 30 that uses intake air pressure and engine speed as parameters, and a predetermined ignition timing value is input therein. In this case, in the ignition timing control learning and correction target area, the ignition timing value at the maximum permissible torque exerted in the internal combustion zone vi, the predetermined fuel used in the internal combustion engine,
For example, for regular gasoline or low-octane gasoline, the limit (knock limit) value of the ignition timing that provides the maximum advance angle that can suppress knocking within an allowable range is used to set each control region. Also R
In the AM 16, learning values are stored in memory for each region divided into operating regions corresponding to the ignition timing II Ill region (region 0 in FIG. 2). Up again! A map 33 (see FIG. 3) is prepared in the aROM 17, and the acceleration determination value (indicated by intake air pressure) is written therein. Furthermore, an acceleration correction map 34 (see FIG. 3) is prepared in the RAM 16 for correction during accelerated driving. This map 34 takes the engine ml rotation speed as a parameter, and writes correction amounts corresponding to each operating range. Next, as a configuration for partially correcting the value of ignition timing, an equicomplementary block circuit configuration corresponding to the microprocessor 18 is shown in FIG. Here, reference numeral 40 is a basic ignition timing setting circuit which becomes a bound book as a result of overall correction, and its output signal S P K tot is sent to a signal 5PKr via a synthesizer.
eal and is used to drive the engine 41. Further, reference numeral 42 indicates a circuit that is equipped with a map (corresponding to the ignition timing control area using intake air pressure and engine speed as parameters) that takes into account the correction amount of the ignition timing, and from here the correction signal is S P K prt is output to a combiner and is combined into the signal S P K tot. Further, reference numeral 43 is a map (times rB
111 when ignited like 42? J (corresponding to the t2IIIi region), and its output signal N U M is connected to the learning curve holding circuit 44 . Knock occurrence interval judgment value circuit 4
Served on 5th. The learning curve holding circuit 44 can selectively extract correction coefficients corresponding to the number of corrections (see Figure 5 (■)), and the knock interval judgment value circuit 45 can selectively extract knock interval judgment values corresponding to the number of corrections. (Figure 5 C>
*See). On the other hand, the knocking that occurred while the engine 41 was operating was
It is detected by the knock sensor 4G and taken into the knock occurrence interval/intensity calculation circuit 47. The signal for the value of the knock intensity is given to a circuit 48 comprising a gain map (retard angle ff1KNK), and the signal KNK extracted from this is:
It is integrated with the correction coefficient LN from the learning curve holding circuit 44,
The value of the corresponding ignition timing correction map is changed by 11 keys. Further, the signal regarding the knock occurrence interval is sent to the comparator 49.
Then, the comparison signal is compared with the output signal (determination value ADJ) from the knock interval determination value circuit 45, and the comparison signal is input to the advance angle value setting circuit 50, and the advance angle 1aADV sets the value of the corresponding ignition timing correction map. It is rewritten. Next, the ignition timing control method according to the present invention will be explained using a flowchart. Under a certain operating condition, step 861.8 determines whether the operating state is a steady operating state or an accelerated operating state.
It is determined by the process of 62.863. In step 561,
Measure the rotation speed and intake air pressure at that time, and proceed to the next step 3.
In step 62, the acceleration judgment value corresponding to the rotation speed is looked up from the map 33 in the ROM 17, and the comparison is performed in step 8G3.
to determine whether the vehicle is in an acceleration state or not. If it is in the acceleration state, the process moves to step 364, and the map 34 in the RAM 16 is
From this, a correction value in the acceleration state is calculated, and the ignition timing value used for control is calculated and output. If the operating state is steady, the process moves to step 865, where the value of the basic ignition timing is extracted from the map in the ROM 17, calculated, the correction value obtained through learning is added to it, and the ignition timing used for ilJ control is calculated. This routine is shown in FIG. If knocking occurs in this operating state, it is determined in step 371 whether or not it is in an acceleration state, and if it is in an acceleration state, the process moves to the acceleration correction execution subroutine 72, and if it is not in an acceleration state, it moves to the next step S73. do. Step 373 is a routine for steady-state ignition timing learning control; here, allJ m PfI
It is the start time, and a flag FTCMP-0 is used to determine whether or not to execute the overall correction. If a total correction is to be performed, the process advances to a total correction execution subroutine S74, and if a partial correction is to be performed, the process proceeds to a partial correction execution subroutine S75. Furthermore, if the value of the microbrossener 18 to be executed after executing the partial correction deviates significantly from the optimum ignition timing, a condition is given in step 876 (whether or not the partial correction amount exceeds the limit value). If the conditions are met, the flag FTCMP is set to O in step 377 in order to return to the overall correction again.
Let's enter the value. After passing through these routines, step Sγ8
Then, the ignition timing 3 P K real to be actually output is calculated. Note that the above step 378 is the same as the above step S64゜S.
It corresponds to 65. The above flowchart is shown in FIG. Next, a routine for correcting the ignition timing when the vehicle enters the acceleration state will be explained with reference to the flowchart of FIG. In step 381, it is determined whether or not knocking has occurred, and if knocking has occurred, a measure for strong knock a' (KNS RG> is executed in step S82, and then as shown in FIG. 3(C) From the relationship shown in
is calculated and added to the correction tri S P K acc extracted from the acceleration correction map 34. The result is then stored as a new correction value in the acceleration correction map 34 in step S89. If knocking has not occurred, it is determined in step S85 whether or not knocking has occurred a specified number of times under the operating conditions. If it has not occurred, the predetermined advance angle 11ADVacc is subtracted from the correction value in step S86, and the correction value S P K ac
Let it be c. In order to prevent the correction value from becoming negative, it is determined in step S87 whether the correction value S P K acc is less than or equal to O, and if it is less than or equal to O, step S87 is performed.
In step 88, the correction value S P K acc is set to O value. Then, in step 889, a new correction WiS P K acc
is added to the acceleration correction map 34. Next, the routine for overall correction of ignition timing in a steady state will be explained with reference to the flowchart of FIG. First, in step 391, it is determined whether or not the current value of each sensor 18 and 4 is within the control range A of the map 30 in the ROM 17. If it is within that range, in step 892, the corresponding operating range ( Checks how many times correction has been performed in one of the areas divided into a grid (designated as shown in FIG. 2(b), for example), and determines whether the correction has reached the specified number of times. . If the specified number of times has been reached '9), at this point in step 893,
The correction value for the relevant operating region is rewritten for the entire region A, and the number of corrections is reset. Then step 394
At this point, a flag FTCMT indicating the end of the overall correction is set to 1. In this way, the correction value of the area in which the number of corrections reaches a predetermined number first among all control areas is used as the correction value for all control areas to predict the overall correction. It is set. Next, the partial correction routine will be explained with reference to the flowchart of FIG. If the current operating state is in the ignition timing 1IIIllI region, first, in step 5101, the engine speed and intake air pressure are calculated, and in the next step 5102, it is determined whether the current operating state is in the control region. . If it is within the range, in step 5103, the correction amount and correction amount aNUM of the ignition fR period up to the previous time are stored in the ignition timing correction amount map circuit 40.15 and the correction number holding map circuit 42.
Look up and calculate from. Then, in step 5104, the correction coefficient LN is obtained from the MW curve holding circuit 44, and the knock seller interval judgment t! 1 circuit 45
From this, the determination value ADJ is determined in accordance with the number of corrections from the correction number holding port 43, respectively. Then, in step 5105, it is determined whether or not there is knocking, and if there is knocking, the knock intensity and knock occurrence interval are calculated in the circuit 47, and the retard angle ff is calculated accordingly.
1KNK is looked up from the gain map circuit 48 (step 5106). In step 5107, this retard angle ff1KNK and the correction coefficient LNfi! ) to calculate the actual retard angle jiRETreal. Then, in step 8108, the previous correction ff1sP
Subtract the retard angle ff1RETreal from Kprt to find the new 5PKprt, and at the time of ignition 111]? It is stored in the main map. If there is no knocking, it is determined in step 8109 whether the knock occurrence interval is longer than the knock occurrence interval (aADJ) (in other words, it is determined whether there was no knocking during a certain FR interval), and whether or not the knock occurrence interval is longer than the knock occurrence interval (aADJ) is determined. Then step 5
At step 110, a certain advance angle ff1ADV is added to the previous correction ff1sPKprt to obtain a new correction amount SPKρrt. Then, in step 5111, this value is set to M
B T -S P K tot (Kokote MB
T is the value of ignition timing at maximum torque, SP K t
ot is the #J# value of ignition timing used for control)
It is determined whether the actual ignition timing is greater than the MBT, and the actual ignition timing is prevented from advancing beyond the MBT. J, big 'a combined ICi;
t, M B T -S P K tot 'c I
iL/IMh positive m SPKρrt, and if it is small, store the new correction l5PKprt in the ignition timing correction amount map (steps 5112 and 3).
113). In this way, the partial correction is completed, but when returning to the main routine, it is checked in step 376 whether large knocking or correction in the same direction continued (how many times the advance or retard angle was performed). The judgment is made, and if this judgment fails, in step S77, the flag F T is set as before.
The CMP is set to O and the entire correction is performed next time. In addition, in the above embodiment, regarding the overall correction, when the predetermined number of corrections for the same operating region is satisfied (for the operating region in which the predetermined number of corrections is reached first), the correction value at that time is set as the correction value for the entire region. However, it goes without saying that the learning condition may not be the number of times of correction, but the number of times of direction changes at which correction to the retard value and correction to the advance value are switched. (Effects of the Invention) The present invention, as detailed above, first applies a correction value with high correction accuracy to the correction value of the entire sub-iXJ region to quickly approximate the optimal ignition timing, and then By making partial corrections, it is possible to converge the ignition timing value in a short time, reduce the non-king order rate, and improve the driving feeling.

[Brief explanation of drawings]

Figure 1 is a block diagram showing one embodiment of the present invention, Figure M2 (
2), @ is a diagram showing the ignition timing control Ill area J5 and correction grid points, the third diagram, (b) is the R used in the acceleration state
Figure 3 shows the acceleration determination map and acceleration correction map in OM and RAM. Figure 3 (C) shows the relationship between the retardation amount and the knock intensity. Figure 4 shows the equi-complementary map for partial correction. Figures 5a and (b> are relationship diagrams for determining correction coefficients, Figure 6 is a flowchart of the ignition timing control execution routine, and Figure 7 is the ignition timing control learning correction execution routine. 8 to 10 are flowcharts showing the same subroutine. 1... Intake air pressure sensor, 2... Buffer, 3...
・4 A/D converters...Engine speed sensor, 5
...Buffer, 6...Interrupt processing circuit, l...Knock sensor, 8...Filter, 9...Amplifier, 1o
... Rectifier/integrator circuit, 11... Amplifier, 12...
Comparator, 13...Input port, 14...Output port, 1s-CPU, 1G-RAM, 17-ROM, 18
...Microprocessor, 19...Output circuit, 21
...Ignition device, 40...Basic ignition timing setting circuit, 4
DESCRIPTION OF SYMBOLS 1... Engine, 42... Ignition timing correction m-map circuit, 43... Correction number holding circuit 77, 44-... Learning curve holding circuit, 45... Knock occurrence interval judgment value circuit, 4G... ... Knock sensor, 47... Knock occurrence interval/intensity calculation circuit, 48... Gain map circuit, 49.
... Comparator, 5o... Advance angle 1 (ADV) set circuit. Patent Applicant Fuji Heavy Industries Co., Ltd. Agent Patent Attorney Makoto Kobashi 4 Patent Attorney Susumu Murai Figure 2 Hoko ζ Figure 3 (b) 34 (C') Figure 5 (b) 繞LE1 [- Figure 6 Figure 7 Figure 8 Procedural Amendment (Method) May 30, 1985 1. Indication of the case 1988 Patent Application No. 280558 2. Name of invention Ignition timing control method for internal combustion engine 3. Make amendments Relationship with the patent case Patent Applicant 1-7-2 Nishi-Shinjuku, Shinjuku-ku, Tokyo 4, Agent 6, Date of amendment order April 30, 1985 (shipped) 6, Drawing plan subject to amendment 7, Contents of correction

Claims (1)

[Claims] In a device that detects the intake air pressure and engine speed and sets the ignition timing value at that time, and corrects the above ignition timing value when knocking is detected, the intake air pressure and engine speed are When the ignition timing value is learned and a predetermined learning condition is satisfied for the same operating region, the entire ignition timing control region is rewritten to the corrected value at that time, and then each An ignition timing control method for an internal combustion engine characterized by correcting an ignition timing value while performing learning for each operating region.