JPS6331668B2 - - Google Patents

Description

[Detailed description of the invention]

The present invention relates to a method for optimally controlling ignition timing for each cylinder in order to improve fuel consumption in an internal combustion engine. The ignition timing of an internal combustion engine is generally determined by an average value that maximizes output and minimizes fuel consumption based on engine speed and intake pressure (or intake flow rate). If the engine has multiple cylinders, it is common to use this average common ignition timing for each cylinder. However, due to manufacturing variations, changes over time, etc., the ignition timing that provides maximum output varies from cylinder to cylinder, and the average ignition timing described above is not always optimal for that cylinder, causing power loss. Become. In view of the shortcomings of the prior art, the present invention provides optimal ignition timing to each cylinder to ensure maximum output efficiency at all times, without being influenced by factors other than ignition timing, such as accelerator operation. The purpose is to enable ignition timing control. To explain the following using drawings, Figure 1 shows the general relationship between ignition timing and engine speed (torque), and shows the optimal ignition timing θopt that sets the engine speed to maximum Nmax with a constant accelerator operation amount. Have one. This optimum ignition timing θopt usually varies between cylinders. Therefore, for simplicity's sake, if we consider a two-cylinder engine, which is the smallest multi-cylinder engine, as shown in Figure 2, when the ignition timing of each cylinder #1 and #2 is changed, the engine speed will be as indicated by the dashed line. The contour lines r ₁ , r ₂ , r ₃ . . .
Therefore, there are combinations of ignition timings θ ₁ opt and θ ₂ opt for the respective cylinders that form the peak of the engine speed,
Conversely, if each cylinder is driven with this optimal ignition timing combination, the maximum rotational speed can be obtained. The present invention uses the method described below to search for a combination of ignition timings for each cylinder that provides the maximum engine speed. First, to explain the principle of this method, in Fig. 2, the ignition timing of the #1 cylinder is set to θ ₁ and the ignition timing of the #2 cylinder is set to θ ₂ , and the operation is performed with this ignition timing combination (point L). Assume that N ₁ is obtained as the engine rotation speed. (Here, for convenience of explanation later, θ ₁ and θ ₂ should be considered as correction amounts for the basic advance angle θ _B determined by the intake air amount (or intake pressure) and engine rotation speed. , the origin in Fig. 2 is not O but θ _B in terms of ignition timing advance, and the actual advance value is obtained by adding θ ₁ or θ ₂ to this.) Next, what is the above point L? A combination of ignition timings at point H where the ignition timing is slightly changed (for example, the ignition timing of the #1 cylinder is slightly increased by Δθ ₁ , while the ignition timing θ ₂ of the #2 cylinder is left unchanged) (θ ₁ +Δθ ₁ , θ ₂ ), and the rotational speed at that time is N ₂
shall be. Furthermore, a combination (θ ₁ , θ ₂ +Δθ ₂ ) at another point S (for example, the ignition timing of the #2 cylinder is increased by Δθ ₂ , while the ignition timing of the #1 cylinder is set to θ ₁ ) Let the number of rotations obtained by operation be _N3 . Then, the rotation speed obtained by driving again with the first combination of points L is
Let N ₁ ′. In other words, the table below shows the ignition timing combinations for a two-cylinder engine.

[Table] Next, find the average value for each cylinder in the combinations of the first three initial ignition timings (i.e., the number of cylinders plus 1), and find the ignition that minimizes the rotation speed due to the influence of ignition timing alone. Search for a combination of timings (in the example of FIG. 2, L for the number of revolutions _N1 ). In this case, the combination L with the same ignition timing is operated twice for a predetermined period, and the rotation speed N ₁ and
The difference in N ₁ ' is considered to be due to a change in rotation speed due to a cause other than a change in ignition timing, such as accelerator operation, and during that time the rotation speed changes linearly between the two rotation speeds in proportion to time due to accelerator operation. shall be.
Then, the rotation speed due to accelerator operation etc. for each ignition timing combination H and S is determined by linear interpolation,
The deviation from the measured rotational speeds N ₂ and N ₃ is determined, and from this deviation a combination L of ignition timings that minimizes the rotational speed due to the influence of the ignition timing alone is determined. Next, on the straight line connecting this average value and the ignition timing combination L at the minimum rotation speed N ₁ ,
A new point _R1 is set on the opposite side to L (in other words, in the direction in which the rotational speed increases). Then, calculate the ignition timing combination (θ ₁ ′, θ ₂ ′) at this point R ₁ ,
Operate with this combination to obtain a rotation speed of N ₄ , and then operate with the ignition timing combination of point H to obtain a rotation speed of N 4.
N ₂ ′ is obtained, and this rotation speed N ₄ is replaced with the minimum rotation speed N ₁ , and from the difference between N ₂ and N ₂ ′, the rotation speed due to accelerator operation etc. at points S and R ₁ is determined by linear interpolation, and this N ₂ from the deviation between the interpolated value and the measured value,
Among the three rotational speeds N ₃ and N ₄ , the point H at which the minimum rotational speed (N ₂ in the figure) is determined by only the ignition timing is determined. Next, the average value of the three ignition timing combinations'
Find this average value' and the point H where the minimum rotational speed is
A new point _R2 is set on the opposite side of the ignition timing combination H that results in the minimum rotational speed (that is, in the direction in which the rotational speed increases) on the straight line connecting the two. If you repeat this, R ₁ → R ₂ → R ₃ → R ₄
→ R ₅ → R ₆ → R ₇ → R ₈ By modifying the ignition timing combination one after another in the direction of increasing rotation speed, you can reach the peak of rotation speed. FIG. 3 shows how to obtain a new ignition timing combination θnew (ie, R) for the three basic points H, S, and L. In this case, the average value is expressed as a vector: =H+S+L/3...(1). Further, if θnew is a new point to be replaced with the ignition timing combination θmin (that is, L) that minimizes the rotational speed, θnew=−α(θmin−) (2) α: constant. For convenience of explanation, we have described two cylinders above, but even in engines with more cylinders, the ignition timing of each cylinder can be corrected and controlled to maximize the rotational speed using the same principle. . FIG. 4 shows changes in ignition timing combinations and variations in rotational speed for a four-cylinder engine. In a, the engine is operated for a predetermined period with each of the above ignition timing combinations and the same ignition timing combination ', and the rotational speed N ₁ of and ' is set.
Combinations of each ignition timing from the straight line connecting N ₁ ′,
, , (rotational speed difference ΔN) is determined, and the combination of ignition timings (ΔN ₃ in the figure) that results in the smallest ΔN is corrected to a new combination of ignition timings'. Next, in b, the engine is operated with the ignition timing combinations , , ′, ′ and the same ignition timing combination ′, and the combination with the smallest ΔN is modified to ′ in the same way as described above, and c
, operate with the same ignition timing combinations as , , , , , and , and correct the combination that minimizes ΔN to . By repeating this operation, the ignition timing combination is optimized It is possible to accurately correct the combination of periods. Note that d indicates how to obtain ΔN by linear interpolation, and ΔN=N−{N ₀ ′−t ₀ ′−t/t ₀ ′−t ₀ (N ₀ ′−N ₀ )}
...(3) is obtained. Next, to explain the apparatus for realizing the method of the present invention, in FIG. ing. Intake air is introduced into each cylinder from an intake manifold 12, and a throttle valve 14 controls the flow rate of this intake air. An air flow meter 16 is provided upstream of the throttle valve 14 to measure the amount of intake air. Note that instead of measuring the intake air flow rate using the air flow meter 16, the pressure inside the intake pipe may be measured. Reference numeral 18 is a rotational speed sensor, which generates a signal according to the engine rotational speed. As the rotational speed sensor, a well-known crank angle sensor that generates a pulse signal at each crank angle of the engine can be used. An ignition device 20 includes an igniter, a distributor, and an ignition coil, and is connected to the spark plug electrode of each cylinder via a wire 22. The ignition control circuit 26 forms the activation signal for the ignition device 20 and functions as a computer programmed to carry out the method described below. The intake air amount sensor 16 and the rotational speed sensor 18 are connected to the control circuit 26 via lines 30 and 32, respectively.
is connected to. The control circuit 26 calculates the ignition timing determined by the combination of intake air amount and rotation speed,
An ignition timing signal corresponding to the result of this calculation is output to the ignition device via line 34. FIG. 6 shows a block diagram of the ignition control circuit 26, whose input port 42 receives signals from the intake air amount sensor 16 and the rotational speed sensor 18. The A/D converter 40 converts an analog signal from the intake air amount sensor 16 (or intake pressure sensor) into a digital signal. Output port 46 serves as a signal gate to igniter 20. The input port 42 and the output port 46 are connected to the CPU 48 and ROM 5, which are the components of the computer.
0, is connected to the RAM 52 via a bus 54, and exchanges signals in synchronization with a clock signal from a clock generator 58. The control circuit 26 is programmed to obtain the optimum ignition timing for each cylinder according to the principle of the present invention described _above . The ignition timings are set at θ ₁ and θ ₂ as shown in A and B, respectively.
By operating in this state, the engine speed changes as shown in C, and an ignition pulse is generated as shown in E. Between the predetermined ignition pulses near the end of this first step _S1
The clock pulses are taken in from Cfo to Cfend and counted, and this is taken as the engine rotational speed _N1 at the first step _S1 . 2nd step
In S ₂ , the first cylinder is kept at θ ₁ + Δθ, the second cylinder is kept at θ ₂ , and the predetermined number of ignitions is performed in the same way, and Cfo ~
Measure the rotational speed N ₂ as the number of clock pulses between Cfend. Similarly, the number of revolutions N ₃ in the third step S ₂
Measure. In addition, in the fourth step _S4 , the engine is operated with the same ignition timing combination as in the first step, and the engine speed is
Measure N ₁ ′. The ignition timings θ ₁ and θ ₂ that result in the minimum rotation speed are calculated based on the average of the rotation speeds measured in this way and the distance from the straight line connecting N ₁ and N ₁ ′, and a new ignition timing is set on the opposite side. The timings θ ₁ ' and θ ₂ ' are determined according to the calculations shown in FIG. 3, and a predetermined ignition operation is similarly performed, and the rotational speed N ₄ is calculated as the number of clock pulses (fifth step S ₅ ). And the sixth step (not shown)
The engine was operated with the same ignition timing combination as in the second step S ₂ , the rotational speed N ₂ ' was measured, and the ignition timing was the only factor among the ignition timing combinations in steps S ₂ , S ₃ , and S ₅ as described above. The combination of ignition timings that minimizes the rotational speed is found using the following formulas, a new ignition timing combination is set on the opposite side, and this process is repeated. The general outline of the program for executing the ignition timing control according to the present invention has been explained above, and the program will be explained in detail with reference to the flowchart shown in FIG. First, when the internal combustion engine starts, the program executes an interrupt processing routine for calculating the ignition timing from step 100. Next, in step 101, the basic ignition timing θ _B is calculated from the intake air amount (or intake pressure) detected by the intake air amount sensor 16 (or intake pressure sensor) and the rotation speed sensor 18 and the rotation speed.
Specifically, the memory stores basic ignition timing mapping for combinations of intake air amount and engine speed, and the basic ignition timing is calculated based on the actually measured intake air amount and engine speed. Step 102
Then, it is determined whether the engine is steady based on changes in the rotational speed and intake negative pressure. If the engine is not steady, branch to NO and go to step 103. 103
Then, set the step counter to 1 = 0, the ignition number counter to Cf = 0, the clock pulse counter to np = 0,
Clear KEY=0 of the flags described later.
Next _, in step 104, the start ignition timing is set for the _{1st , 2nd} , _{3rd , .} Set θ _B + θ _M. Here, θ ₁ , θ ₂ ... θ _M are ignition timing correction amounts, which are stored according to the intake air amount (or intake pressure) and rotation speed, as explained with reference to FIG.
The ignition timing is calculated by adding the basic ignition timing calculated in step 101 to this. At step 105, the program returns to the main routine. If it is determined in step 102 that it is steady, the branch goes to YES and the ignition timings ₁ , ₂ , . . . set as above are set.
Operation is performed using combinations _M , and the rotational speed is measured at the first point L in FIG. 2 (ie, the first step S ₁ in FIG. 7). First, in step 106, the value of counter J, which is cleared for each number of cylinders, is incremented by one for each ignition, and in step 107, it is determined whether the value of J has reached the number M of cylinders. If the value of J has not reached M, clear it to J=1 in step 108,
If it has not been reached, proceed to step 109. In step 109, the ignition timing _J is determined for each cylinder by adding the basic advance angle θ _B and the ignition timing correction amount θ _J . In step 110, the ignition number counter Cf
is incremented by one for each ignition. 111
In step , it is determined whether the number of ignitions Cf is Cfend in FIG. 7(e). At first, it naturally branches to NO, and in step 117 it is determined whether or not the number of ignitions Cf is greater than Cfo in FIG. If NO, this indicates that the rotational speed measurement period has not yet begun, and the process returns to the main routine at step 119. At 117, the number of ignitions Cf
The number of firings at which clock pulses should begin to be measured.
If it recognizes that Cfo has been reached, it branches to YES,
After starting the clock pulse count at 118, the program returns to the main routine at 119. If it is determined in step 111 that the number of ignitions Cf has reached Cfend, then in 112 the clock pulse count value np at this time is stored in the memory. This count value represents the engine rotational speed _N1 at step _S1 . And at 113, set the ignition number counter to Cf=
0, clear the clock pulse counter to np = 0, add one to the step counter, set the next ignition timing, operate at point H in Figure 2, and proceed to the second step in Figure 7. Enter _S2 . First, in 114, it is determined whether or not i≦M.
At this time, since i=1, it branches to YES, and in step 115, the ignition timing correction amount for each cylinder is: θ ₁ = θ ₁ + Δθ ₁₁ θ ₂ = θ ₂ + Δθ ₂₁ …… (5) θ _M = θ _M +Δθ _M1 is set. (In Figures 2 and 7, in the second step, the ignition timing of the #1 cylinder is moved by Δθ ₁₁ =Δθ, and the #2 cylinder is explained as is, that is, with Δθ ₂₁ = 0, but Δθ ₂₁ is not 0. ) Return to the main routine in step 116. When the interrupt is entered again from step 100, the engine speed at this second step _S2 is determined by the steps 106, 107, 109, 110, 111, 117, 118, and 119 as in the first step.
N ₂ is measured as the number of clock pulses and the result is
Stored in memory in 112 steps. after that,
At step 113, the ignition number counter Cf and clock pulse counter np are cleared, and the step counter i is incremented by one, and the program enters the third step. First, 114 still determines YES even in the case of two cylinders, and 115 determines the ignition timing correction amount as θ ₁ = θ ₁ + Δθ ₁₂ θ ₂ = θ ₂ + Δθ ₂₂ …… (6) θ _M = θ _M + Δθ _M2 Set. After that, after returning to the main routine at 116, this program is interrupted again at 100, and the processing from 106 onwards is carried out as before, and the engine rotation speed _N3 at this third step _S3 is measured in the form of the clopass count value np. and store it in memory at 112. In the next step 114, it is determined whether i≦M.
In the case of two cylinders, the process branches to NO at this point, and in step 120 it is determined whether the flag KEY=0.
In this case, since KEY = 0, the process proceeds to step 129 and sets KEY = 1. In the next step 130, the amount of correction is the same as that of the first combination of ignition timings whose revolutions are compared (in this case , step
104, θ _{1 ,} . Next, when step 120 is reached again, KEY=1
Therefore, it branches to NO, and after setting KEY=0 at step 121, the rotation speed change ΔN for each ignition timing combination explained in Fig. 4 is calculated as (3) at step 122.
Calculated from the formula. In step 123, the minimum step i=min among the rotational speed changes ΔN is determined. Next, in step 124, the average ignition timing is calculated using equation (1): =H+S+L/3. Since this equation is a vector representation, it is calculated for each cylinder component in each equation (4), (5), and (6). Then, in step 125, a new ignition timing θnew for increasing the rotational speed is calculated based on this average ignition timing using the second equation θnew=-α(θmin-). Since this equation is also expressed as a vector, the new ignition timing is calculated for each component of equations (4), (5), and (6). Next, in step 126, this new ignition timing θnew is used as a combination of ignition timings θmin (i.e., θ ₁ , θ ₂ , . . . in equation (4)) that constitutes the minimum rotational speed change.
θ _M ). Then, in 127 steps, the ignition timing correction amount is set for each component, and θ ₁ , θ ₂ ,...
..., θ _M is obtained, and the process returns to the main routine at step 128. Once the new ignition timing combination is set like this,
At the next interruption, the fifth step in FIG. 7 is executed by the routines 106, 107, 109, 110, 111, 117, and 118 in FIG. ₈ , and the rotational speed N4 is measured.
After that, it branches from step 111, passes through 114120129, and at 130, the same correction amount as in the second step is set and operation is performed, and the rotation speed N ₂ ' is measured.
Then, as described above, the combination of ignition timings with the minimum rotational speed change ΔN is replaced with a combination in the direction of increasing the rotational speed. Thereafter, the same process is repeated to accurately determine the peak of the rotational speed. As described above, the present invention sets a plurality of ignition timing combinations for each cylinder, performs predetermined calculations to correct the ignition timing combinations so that the engine speed increases from these combinations, The method is to obtain the combination of ignition timing for each cylinder that maximizes the engine speed in each operating state. In particular, operation with at least one ignition timing combination is performed twice, and the rotation speed, etc. in these two operations is determined. Since the operating status signal for each ignition timing combination is corrected based on the difference in the operating status signal, it is possible to control the optimal ignition timing without being affected by external factors such as accelerator operation, and the output efficiency is always maintained. This has the excellent effect of ensuring a maximum of .

[Brief explanation of the drawing]

Fig. 1 is a characteristic diagram showing the relationship between ignition timing and engine speed, Fig. 2 is a contour chart showing changes in engine speed for ignition timing combinations in the case of two cylinders, and Fig. 3 is a diagram showing the change in engine speed with respect to the ignition timing combination in the case of two cylinders. Figure 4a, b, and c are diagrams showing changes in ignition timing combinations, and Figure 4d is a diagram showing how to calculate the engine speed. A diagram showing the method, FIG. 5 is a configuration diagram of an internal combustion engine according to the present invention, FIG. 6 is a configuration diagram of the control circuit in FIG. 4, and FIG. 7 is a timing diagram showing the concept of arithmetic processing in the present invention. , FIG. 8 is a flowchart showing the ignition timing calculation processing procedure in the present invention. 10... Engine body, 16... Intake air amount sensor, 18... Engine rotation speed sensor, 20...
Ignition device, 26...control circuit.

Claims (1)

[Scope of Claims] 1. A plurality of combinations of ignition timings having a predetermined value for each cylinder near the target ignition timing are selected, and each combination of the selected ignition timings is sequentially operated for a predetermined period of time. Detect operating state signals such as engine speed during each operation, and compare the detection signals for each of the above ignition timing combinations to find the ignition timing combination that results in the operating state farthest from the optimal operating state. , an ignition timing control method that changes the ignition timing combination to a new ignition timing combination to modify the ignition timing combination in a direction that provides an optimal operating condition, the ignition timing control method comprising: changing the ignition timing combination to a new ignition timing combination; The operation with at least one combination of ignition timing is performed twice, and these two
Determine the difference between the operating status signals during the two operations;
An ignition timing control method for a multi-cylinder internal combustion engine, characterized in that the operating state signal for other ignition timing combinations is corrected based on this difference. 2. The ignition timing control method according to claim 1, wherein the two operations are performed at the beginning and the end of the operations in the selected combination of ignition timings.