JPS6243056B2 - - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION The present invention relates to an engine control method, and in particular,
Suitable for use in automobile engines equipped with a microcomputer, it is suitable for adjusting correction values for engine control target settings such as ignition timing and air-fuel ratio.
The present invention relates to an improvement in an engine control method in which the memory is stored in correspondence with the engine operating state, and this memory is updated in accordance with the state of engine control factors such as knocking and excess air ratio detected by a control factor sensor.

With the recent worsening of the oil situation, automobile fuel efficiency has become a social issue. One possible method for improving fuel efficiency is to improve engine efficiency, and one specific method is knock feedback control of ignition timing. This is because, in general, engine knocking is a dangerous phenomenon that can lead to engine destruction in the worst case, so with conventional ignition systems, the ignition timing is adjusted by setting a certain amount of margin from the ignition timing at which knocking occurs. By detecting the knocking state of the engine and performing feedback control on the ignition timing, the advance margin from the point of knocking occurrence, which had to be taken into consideration with the conventional technology, is increased. This eliminates fuel consumption and improves fuel efficiency accordingly.

However, the disadvantage of general feedback control, including this knock feedback control, is that it is difficult to improve transient response because some result is produced and the cause is corrected for that result. The control accuracy during steady state cannot be increased beyond a certain level. Therefore, even in this knock feedback control, several knocks always occur during transient periods, and even if the steady state continues for a long time, the knocks do not disappear completely.

The same thing can be said about air-fuel ratio feedback control using an excess air ratio sensor. This control is widely used from the perspective of exhaust gas regulations, but this control also has problems in terms of transient response and control accuracy.

As shown in Japanese Patent Application Laid-Open No. 54-112, one way to partially eliminate these drawbacks is to always store the latest information held by the engine in correspondence with engine operating conditions such as engine speed and intake pipe negative pressure. while doing,
It has also been proposed to overcome the control delay in engine combustion control by controlling factors that affect combustion, such as the air-fuel ratio, ignition timing, and exhaust gas recirculation rate, based on the stored contents.

By the way, the information stored in this way also needs to be updated from time to time in order to respond to changes in the environment. However, if this update is performed without taking into account the previous information, information that takes into account transient factors at the time of update will be stored, and repeated updates will accumulate experience. However, it is important to perform highly accurate control that takes advantage of experience, and to perform updates by giving sufficient weight to the information before updating so as to make use of past experience. The influence of information persists even after repeated updates, resulting in poor control responsiveness.

In view of the above problems, an object of the present invention is to store a correction value for a set value of an engine control target in correspondence with the engine operating state, and to store this memory as an engine control factor detected by a control factor sensor. An object of the present invention is to improve control accuracy in an engine control method that updates according to the state of the engine.

For this reason, the present invention calculates and stores the number of updates, and when updating, weights the correction value before the update according to the number of past updates, and calculates the correction value after the update. It is something.

According to the present invention, when updating the correction value, the correction value before the update is made use of according to the number of updates, so when the number of updates is small, that is, while there is little experience, the correction value is updated relatively large and the correction is performed. On the other hand, after the number of updates increases and experience is gained, the correction value is updated only little by little, making it possible to perform stable control with high control accuracy by making use of the experience up to that point.

Hereinafter, the principle of the present invention will be explained in the case where the engine control target is ignition timing and the control factor is knocking. The present invention is based on engine control using a microcomputer. First, an ignition timing considered to be optimal under standard engine conditions is determined and stored in a memory element as a set ignition timing. Specifically, it is advantageous to configure the storage element with a rewritable nonvolatile memory so that the set ignition timing can be changed depending on the type of engine. Normally, feedforward control is performed using this set ignition timing (solid line A in Figure 1), but if, for example, the gasoline used changes and knocking becomes more likely to occur, the required optimal ignition timing In the case of gasoline B, it is as shown by the broken line B in Figure 1, and in the case of gasoline C, the first
It becomes as shown in the dashed line C in the figure. The amount of correction from this set ignition timing to the optimum ignition timing (Fig. 1D)
In normal knock feedback control, this is changed each time, but in the present invention, this correction amount is learned from the actual knocking occurrence situation and stored separately from the set ignition timing, thereby improving transient response. At the same time, we are trying to suppress variations during steady state. Specifically, the storage element for this correction amount is composed of a rewritable volatile memory,
Is the size the same as the amount of memory of the set ignition timing?
Otherwise, it is considered to be the range where knocking occurs.

FIG. 2 is a block diagram showing an engine control device for realizing the present invention. The engine 1 includes a crank angle sensor 2, an intake pipe negative pressure sensor 3,
The output signal of the crank angle sensor 2 is inputted to the counter 6 along with the output of the clock pulse oscillator 5, and the output is inputted to the register 9 as the engine rotation speed. Further, the output signal of the intake pipe negative pressure sensor 3 is input to an analog-to-digital converter 8, and the result is input to a register 10. The registers 9 and 10 are connected to a microcomputer 12a, and the microcomputer 12a includes read-only memories (hereinafter referred to as ROM) 13a and 13b for computer operating programs and ignition timing settings. The output signal of the knocking sensor 15, which detects engine knocking from the vibration of the engine body or the sound waves generated by this vibration, is sent to the microcomputer 12 via the knocking determination circuit 17.
b. Also, the microcomputer 12
b is a random access memory (hereinafter referred to as RAM) 14 for storing correction values for ignition timing;
RAM for storing a and the number of updates of correction values
14b. Microcomputer 12
A and b are configured as a single computer and can exchange data. The output of this microcomputer is connected to the ignition device 19. Here, ROM13b, RAM14a,
14b shows one method of storage. Figure 3A is an ignition timing setting map in which set ignition timing is stored in correspondence with engine speed and intake pipe negative pressure, Figure 3B
The third ignition timing correction map stores only the correction amount in correspondence with the engine speed and intake pipe negative pressure.
Figure C is an update number map in which the number of updates of the correction amount is stored in correspondence with the engine rotational speed and the intake pipe negative pressure. The ignition timing setting map, ignition timing correction map, and update frequency map are all created according to the engine speed N and intake pipe negative pressure Pb, and the ignition timing and frequency are entered at each grid point. . Conditions other than grid points are determined by interpolation. In addition, the values in the ignition timing correction map are:
It shows the amount of retardation learned by knock feedback control.

Now, the operation of the apparatus shown in FIG. 2 will be explained with reference to the flowchart shown in FIG. According to the program stored in the ROM 13a, first, engine speed N and intake pipe negative pressure Pb are read from registers 9 and 10 as basic control factors for engine control.
From this value, the optimum set ignition timing θp under the standard engine operating condition is calculated from the ignition timing setting map as shown in FIG. 2A, which has been previously stored in the ROM 13b.

At this time, as mentioned above, points other than the grid points can be found by interpolation, and if the increments of the map are small,
A grid point may be used as the representative point. It is also possible to obtain it as a function from the following equation instead of using a map.

θp=f(N, Pb) ...(1) Next, detect whether or not knocking is occurring in the current engine condition from the output of the knocking sensor 15, and adjust the knock feed according to the result. Determine the necessity of back control. If it is determined that knocking has not occurred, the correction value _θ The final ignition timing θig is calculated using the following formula.

θig=θ _P −θ _F (2) Based on this calculated ignition timing, the ignition device 19 performs ignition. If it is determined that knocking has occurred, a new correction must be made to the ignition timing correction map in the RAM 14a. Therefore, the knocking determination circuit 17
Then, determine the strength of Notsuking, and from that information,
Calculate the additional correction request amount (retard amount) θ _T of the ignition timing. This retard amount θ _T is calculated depending on the frequency of occurrence of knocking, for example, and when the frequency of occurrence of knocking is high, the amount of correction required for one time is large, and when the frequency of occurrence is low, the amount of correction required is once. The amount of correction requested is small. The retard amount θ _T calculated in this way is temporarily stored in the computer 12b, and then, like the ignition timing setting map and ignition timing correction map described above, it is calculated from the update number map stored in the RAM 14b. The number of updates k of the ignition timing correction map up to now is calculated. Once the number of updates k has been calculated, the update coefficient p is calculated based on this using the following equation. However, a is a constant.

p=e ^-ak (3) An example of the change in the update coefficient p with respect to the number of updates k is shown in FIG. However, when the number of updates k is zero, the update coefficient p is 0.75 regardless of equation (3).
shall be set to . Next, the ignition timing correction map is updated using the calculated retard amount θ _T and update coefficient p. The calculation for this update is performed using the following equation, and the correction value of the ignition timing correction map in the RAM 14a is updated to the calculated correction value.

θ _F =θ _F ^-1 +p·θ _T (4) However, θ _F ^-1 is the correction value before updating. Next, the number of updates is calculated using the following formula, and RAM1
The update number map of 4b is updated. however,
k ^-1 is the number of previous updates k=k ^-1 +1 (5). The subsequent action is similar to the case where knocking does not occur, as described above, with the correction amount θ
_F and final ignition timing Qig are calculated, and ignition is performed based on the results.

As described above, according to the present invention, when updating a correction value for a setting value such as ignition timing, the correction value before update is weighted according to the number of past updates, and each time the correction value is updated. This prevents the system from being updated completely, and increases the weight of the correction value before the update as updates are repeated, making full use of past experience and greatly improving control accuracy. can. FIG. 6 shows this, and the normal knock feedback control shown in FIG. 6A is performed once (FIG. 6A), twice (FIG. 6B), and in accordance with the present invention. It is clear that the responsiveness of the ignition timing control to the throttle change and the control accuracy in the steady state improve as the number of times (FIG. 6C) increases. Therefore, it is possible not only to improve fuel efficiency and exhaust purification performance, but also to ensure driving performance.

In the above explanation, an example has been described in which the method of the present invention is used for ignition timing control, but the method of the present invention can also be used for air-fuel ratio control, etc.

[Brief explanation of the drawing]

FIG. 1 is a diagram showing the relationship between engine speed, set ignition timing, and optimal ignition timing when the quality of gasoline changes, and FIG. 2 is a diagram showing engine control for realizing an embodiment of the present invention. 3 is a block diagram of the device; FIG. 3 is a diagram showing an example of the ignition timing setting map, ignition timing correction map, and update number map; FIG. 4 is a flowchart showing the flow in the embodiment of FIG. 2; The figure is a diagram showing the relationship between the number of updates and the update coefficient, and FIG. 6 is a diagram showing the relationship between the number of trials and the ignition timing change state in an embodiment of the present invention. 1...Engine, 2...Crank angle sensor, 3
... Intake pipe negative pressure sensor, 5 ... Clock pulse oscillator, 6 ... Counter, 8 ... Analog-digital converter, 9, 10 ... Register, 12a, 12
b...Microcomputer, 13a, 13b...
...Read only memory, 14a, 14b...Random access memory, 15...Knocking sensor, 17...Knocking determination circuit, 19...Ignition device.

Claims (1)

[Claims] 1. A correction value for a setting value of an engine controlled object is stored in correspondence with an engine operating state, and this memory is updated according to a state of an engine control factor detected by a control factor sensor. In an engine control method, the number of updates is calculated and stored, and when updating, the correction value before the update is weighted according to the number of past updates, and the correction value after the update is calculated. An engine control method characterized by: 2. The engine control method according to claim 1, wherein the update of the correction value is calculated by θ _F =θ _F ⁻¹ +pθ _T , where p is a coefficient that decreases with the number of updates. Here, θ _F is a correction value after update, θ _F ⁻¹ is a correction value before update, p is an update coefficient, and θ _T is an additional correction request value for the correction value before update. 3. The engine control method according to claim 2, wherein the update coefficient is calculated by p=e ^-ak . However, a is a constant and k is the number of updates.