JPS62233451A - Control device for internal combustion engine - Google Patents

Abstract

PURPOSE:To make it possible to optimumly control an internal combustion engine, by changing the control gain of proportional-plus-integrating control in accordance with the result of discrimination between a steady-state operating condition in which variations in the rotational speed and the position of a rack is small and a transient operating condition in which variations in the rotational speed or the rack position is large. CONSTITUTION:In a fuel injection pump 1, a fuel injection amount regulating rack is controlled by a metering rack control actuator 2 which is controlled by a control section 21 in accordance with an actual rotational speed detected by a rotational speed sensor 6, a set rotational speed obtained from the output of an accelerator position sensor 7 and a set rack position obtained from the output of a rack position sensor 4. In this arrangement the control section 21 discriminates between a steady-state condition in which variations at least in the rotational speed and the rack position are small and a transient condition in which variations in the rotational speed or the rack position is large, and with the result of the discrimination the control gain of proportional-plus- integrating control is changed.

Description

[Detailed Description of the Invention] Industrial Application Fields The present invention relates to internal combustion engine control '3A'fl, and in particular, a control rack of a fuel T8F pump is driven by a solenoid actuator. 7゜Regarding injection failure control in a control device <Prior art> A control device using a microcomputer etc. controls the actuator for driving the metering rack of the fuel injection pump, controls the position of the metering rack, and controls the fuel injection. It is well known to control the
Type 11 is used (Japanese Patent Application Laid-Open No. 60-256529
(see publication).

In the IR state of the engine.

(,-) Steady operation that controls minute displacements at high speed.

(LI) Transient operation to control one large change at high speed, (c
) During starting to control large displacements with high precision.

However, due to the characteristics of the solenoid, the rate of change in attraction force with respect to the rate of change in current is not constant, and the control gain when moving the solenoid in IJ by proportional-integral control is not the same depending on each of these states. Conventionally, therefore, it has been common practice to repeat tests and select a gain that satisfies each requirement.

<Problems to be solved by the invention> 1. The control gains selected as described above are not necessarily optimal for all operating conditions;
) With the characteristics suitable for (b), actuator hunting is likely to occur in the case of (c) <, with the characteristics suitable for (b), the control accuracy of (, 1) is insufficient, In terms of characteristics, (, +) (b) is likely not to be satisfied, and there is a problem that it may not be possible to appropriately respond to various operating conditions with nine different conditions.

Examples of such cases are shown in FIGS. 2 and 3.

Figure 2 shows the case where the control gain is optimal for a large amount of change.
With respect to a change of about 5 to 10 nin in el, the actual Rac j of the rack position [ ] follows with a delay of about 50 m5ec.With this control gain, the set value Rsat is changed to, for example, !'1m5cc. If the distance changes by about 0.1 m++, the actual value 1jacL cannot follow up sufficiently as shown in Figure (b).Furthermore, Figure 3 shows the case where the control gain is optimal for a small amount of change. (d) As shown in the figure, the set value Rset of the rack position decreases by about O, l nu at intervals of !"in+scc, whereas the actual rack position 11i'j Rrr e 1. quickly follows. If the control gain changes the fixed value R8el by, for example, 5 to 10 uia, the actual value RacL becomes (
b) As shown in the figure, hunting occurs because it changes too rapidly. .

The present invention has been made to address these problems and to provide a control device for an internal combustion engine that can appropriately respond to each driving condition of 1 m.

<Means for Solving the Problems> J. In order to achieve the above objects, the first Tsubakiaki of the present application proposes a system in which a metering rack for adjusting the injection pressure is driven by a solenoid-type rack actuator. The fuel injection pump
Rack position detection means for detecting the position of the upper and lower rack; engine rotation speed detection means for detecting the actual value of the engine rotation speed; and accelerator position detection means for detecting the setting field of the rotation speed of the engine 1 based on the accelerator position; 11 of the rack position depending on the actual value and set value of the detected rotational speed and the rack position.
A control means for calculating the drift value and outputting a control signal for driving the solenoid-type rack actuator by proportional-integral control; It is equipped with a function to distinguish between a steady operating state with a small change in rack position and a transient operating state with a large change in the number of revolutions or a large change in rank position. The control gain of the integral control is changed.

Also, the second light is similar to the first invention.

1. A fuel injection pump in which a solenoid-type rack actuator drives a nozzle (1 week V° laso' for V-adjustment); 1. rack position detection means for detecting the position of the starting μ rack, engine speed/engine speed detection means for detecting the actual value of the engine speed, and accelerator position detection T for detecting the set value of the engine speed based on the accelerator position; and a control means for calculating a target value of the rack position according to the detected actual value and set value of the rotation speed and the rack position, and outputting a control signal for driving the solenoid-type rack actuator by proportional-integral control; It is equipped with The control means is provided with at least a function of discriminating between a steady operating state and an idling operating state, and changes the control gain of the proportional-integral control according to each operating state.

<Operation> Since the control device for an internal combustion engine of the present invention has a function of determining the operating state in the control means and changing the control gain, in the first invention, the internal combustion engine control device has a function of determining the operating state and changing the control gain.
And starting operation condition 1r? if necessary. , etc., and in the second invention, the control gain automatically changes depending on each operating state such as a steady operating state, an idling operating state, and, if necessary, a starting operating state. And depending on the driving condition at that time! Control is performed using suitable characteristics.

<Example> The embodiment shown in FIG. 1 will be described below.

Figure 1 is a conceptual system: diagram a, where (1) is the fuel injection pump, (2) is the %l rack actuator for fuel metering m, and (1) is the timer rack actuator for adjusting the injection timing. (F) (5) is the position center for each of the 7 actuators, (6) is the rotation speed sensor, (7) is the accelerator position, (8) is the accelerator, (9) is the cooling water temperature sensor,
(10) is an intake air temperature sensor, and (11) is a fuel temperature sensor.

(12) is the battery, (1:31 is the key switch.) The main body of the engine is not shown.

A linear solenoid type actuator is used for the adjustment actuator (2).

7 The position of Tachiyuator (2) is 1, for example, the difference is 1? ] I.
It is detected by the position sensor (4) from the lance. In addition, detecting the engine rotation speed 1.14 For example, the movement of the concave groove (17) of the 62-way rotating body (1G) attached to the camshaft (15) is 71
! Become more familiar with the detection by the rotation speed sensor (G), which has a magnetic pin of 7 or 1.

(21) is a control unit that controls the operation and magazine status of the engine according to instructions from the operator.The second control unit (21) is configured using, for example, a Markcro computer, and receives various analog signals. Rero analog input port (22),
A multiplexer (23), an A/D converter (24), a digital input port (25) into which digital signals are input.
) 1 Waveform into which the signal from the rotation speed sensor (6) is input: the pocket circuit (2G), the timer circuit (27), the counter (2
8), RAM used for various control calculations (2!J)
, a ROM (301) that stores control programs and various control data, and a metering rack actuator drive circuit (
3]), timer rack actuator'yJA111J
A circuit (32), etc. are provided, and each of these circuits is connected to the 11erroCr'1l (33) for various control calculations and input/output instructions.

The ROM (30) stores data for calculating the target value of the rack position according to the detected actual value and set value of the rotation speed and the actual value of the rack position. Each operating state is stored in the form of a formula or a numerical table (map) for each operating state, such as starting operating state, idling operating state, etc. in **.

Next, the overall operation will be explained with reference to a series of flowcharts 1. shown in FIGS. 4 and 5.

For control, the ON signal of the key switch (13) is input to the input board (
25) is input and stars 1~. First, after clearing old data, coolant temperature Tw and fuel temperature Tf.

Recognizing the intake air temperature 1゛1, etc., and also recognizing the voltage coefficient MVB of the battery (12) L2, from each map mentioned above, calculate the rack position Rss for 1 mark during start control and the 1 mark for start control.
Imarac position [Tss is determined. Until the starter is driven, both the operating state flag ive and the engine stalling state flag 5tool are 0, and the process continues at step S.
1 and S2, set the metering rack position to r111 Rss, and set the timer rack position to Ts! 1,
CIIU (33) to metering rack actuator F12
N+ circuit (31) and timer rank actuator UA
Control (No. n QouL and Tout are output to the subcircuit (32). In this way, the metering rack actuator (2) and the timer rack actuator (3) are driven to a predetermined FL in preparation for starting the first engine. Note that this starting preparation This operation is possible when electric actuators (4) and (5) are used instead of hydraulic types, which do not operate unless one engine is operated.

In this start preparation state, the pre-start timer PL, ime is set to 1-, and if the starter is not driven by the pre-start timer end time Pend, the process advances to step S3 and the start [・
The status flag 5tool becomes 1, the output Qout to the metering rack actuator (2) and the output TouL to the timer rack actuator (3) are turned off, and the start preparation is terminated. The above pre-start timer end time Pe
nd is set to jfl, for example, about 4 to 10 seconds. On the other hand, if the starter is driven by the end time r'a++d of the pre-start timer, the lotus rotation state flag becomes l, and the fifth
Proceed to figure.

First, the actual value N + l e l of the engine speed is recognized, and the environmental conditions such as the cooling water temperature Tv, the fuel temperature Tf, the intake air temperature Ti, and the voltage coefficient MV[1 of the battery (12)] are recognized again. . Start recognition flag SL, OK starts at 1 rotation speed NacL! 1! When the recognized rotational speed NSL or higher, the signal becomes 1, and while the rotational speed at the initial stage of startup is low, the process proceeds from step S6 to step S87, and the target values Rss and ras during startup control are set as the metering rack position and timer rack position.
are used, and driving by the starter continues while the numerical values are updated sequentially as the rotational speed increases. The above-mentioned starting recognition rotation speed Nsj is a reference rotation speed at which it is determined that one engine has reached a complete explosion state and the engine has started. When this rotation speed is reached, the start recognition flag St and OK are set in step S8.
becomes 1, the start control ends, and the process goes to step S9.

Ens1 - Recognized rotational speed N5 Lool is rotational speed N; +c
When t is below this level, it is determined that an engine stall has occurred.In this case, the operating state flag Drive is set to O, and the engine stall state flag 5 Lool is set to 1.
Returning to FIG. 3, the output to each actuator is turned off in steps S4 and S5.

On the other hand, if the number of revolutions per revolution Nacj is larger than Nd, ool, the process proceeds to step 10, where the set value N5el of the number of revolutions is recognized from the position of the accelerator (8), and the rack position 11vi Rs e is set to the target tone during steady operation. Determined as t; evening when the train is running (
Steady operation is performed using the marac position Pf Tsct;

1. Steps Sll and 12 indicated by 1L broken lines in FIGS. 4 and 5 are subroutines for fault diagnosis, and are inserted as necessary.

Next, an embodiment of the first invention will be described with reference to FIGS. 6 to 9. 6 to 8 are a series of routines, in which the metering rack TypH! step Sl in FIG. 4 and step S7 in FIG. 5 are shown. This figure shows the details of the lI routine.

First, in step S21, it is determined whether normal operation or start control is in progress based on the start recognition flag SjO. If the start recognition flag is in operation, the acceleration recognition flag Nupc and the deceleration recognition flag Ndn are used.
c, it is determined whether the operating state is a steady operating state or a transient operating state during acceleration/deceleration (recognition of acceleration/deceleration will be described later with reference to FIG. 9).

If each flag is 0, it is in a steady state, so step S
Proceed to step 22 and calculate the constant loss state actuator proportional gain GP.
The transient state actuator proportional gain (GvP) is compared, and when GP≧GvP, the difference between Rse and the actual rack position is compared with the hunting recognition rack position Ii'? error RaekOV. 6
If the difference is small, the difference between N5eL and N;icl: is compared with the gain change rotation error N1dlOK, and Rs+>L
and 1(act) is the rank position ['
? It is compared with the error RackO, and if the difference is small in both cases, the process proceeds to step S24, and if the result is negative for any chair 7, the process proceeds to step S25. Since this is a steady operation state, the actuator proportional gain Ga1nP and the actuator integral gain G; +i
As for nl, only slight corrections are made by adding +α or -α to the previous value, and (4) and t? in FIG. 7 respectively.
Proceed to 4). On the other hand, if the result in step S22 or S23 is negative, the process advances to step 826.

Further, if the deceleration recognition flag Ndnc is 1, it is determined that deceleration is in progress, and the process proceeds to step 327 and subsequent steps, and the Nse and Nac
The difference between L is N1dlOK, Rsel, and R;+cL
The difference between them is expressed as RackOV and RackO, respectively, and the difference between the current N*cL and the previous rotational speed NacL(
1) is compared.

If the difference is small in either case, or if the current rotational speed is smaller than the number of turns, the process advances to step 328. In this case, since the deceleration is gradual, Ga1nP and (, >jnl are assumed to be the steady state actuator proportional gain fGP and the steady state actuator integral gain (8).

Proceed to 4) in Figure 7. Further, if the result is negative in either case, the process advances to step 326.

Further, if the acceleration recognition flag Nupc is 1, it is determined that acceleration is being performed, and the process proceeds to step S26. This step S26 corresponds to a transient operating state, and Ga1n P and q
Let ainI be the transient actuator proportional gain GvP and the transient actuator proportional gain GvI and proceed to the plane of FIG.

Moreover, if the start recognition flag 5LOK is not 1 in step S21, one engine is under start control, so (C in FIG. 7)
Proceed to.

Next, in step S29 in FIG. 7, Ga1nP is
Then, Ga1nl is set to 1-1 after GvT, and in step S30, G31n P k G P
Below, Ga1nI is set to be less than or equal to Gl, and in step S31, GaxnP is set to the starting state actuator proportional gain Pst, and Ga1nl is set to the starting state actuator integral gain Ist.

Proceed to the 8th session.

Figure 8 shows the procedure for calculating the actual output Qout to the h1 rutter actuator (2). First, R; J C
l.

, and calculate rack position error integral 11 using Gajn [
1j RC1 is obtained, and in step S32, the control output 1+fiQout' is obtained using Gd1nP. Then, if the start recognition flag SL, 0 is 1, QouL' is corrected by the voltage coefficient M V n of the battery (12), and in step S3:3, the actual output Q (J u
1. Decided to stay. The voltage coefficient MV[3 is the ratio of the base voltage to the actual voltage and is used to compensate for the slow operation of the actuator when the voltage is low.

l: QOUj' in the following corresponds to the duty to operate the actuator (2), and Qo
It corresponds to the duty pulse width according to the voltage coefficient MVB. Ru. And this output Qo
Each driving letter is in uL! Since the control gain is used according to the order, the injection is suitable for each operating condition. )l@sub control will be performed.

FIG. 9 shows the procedure for outputting the acceleration recognition flag Nupc and deceleration recognition flag Nd+u2 used for determining the driving state in FIG. 6, and N5e in step S10 of FIG.
This is done in step jj93fi. here.

First, the engine speed setting IcLNseL; is compared by going back three times. When the set value is increasing, the acceleration recognition flag Nupc is set to 1 if the difference is greater than or equal to the acceleration recognition reference value Nup, and when the set value is decreasing, the difference is greater than or equal to the deceleration recognition base 1 false value Ndown. If so, set the deceleration recognition flag Ndne to 1.

If there is no difference between the set values or the difference is smaller than the reference value, it is determined that the accelerator is not operated, so both the acceleration recognition flag Nupc and the deceleration recognition flag Ndnc are 0.
It is assumed that it will remain as it is.

Next, the second invention will be explained in detail.

While the control of the first invention is performed with very high precision, this invention changes the control gain depending on the steady operating state and the idling operating state when such precision is not required. This was done to simplify the rough hardware and reduce costs.

That is, FIGS. 6 and 7 in the embodiment of the first invention
Instead of the procedure shown in the figure, the procedure shown in FIG. 10 is performed. In this embodiment, first, it is determined whether or not the startup recognition flag S1; UK is in the startup operation state, and if the flag is 1, step S
At 40, a determination of the idle operating state is made. and,
If it is not in an idling operating state, it is assumed that it is in a steady operating state, and Ga1nP is the steady state actuator proportional gain GP.

G, JlrII is the steady state actuator integral gain C
It is considered as I. Also, if it is in idle operation, Ga1n
P is the transient actuator proportional gain GvP,
Ga1nl is the transient state actuator integral gain Gvl
It is said that On the other hand, in the case of starting) i1 rotation state, Ga1
nP is above the starting state actuator proportional gain Pb, and Ga1nl is the starting state actuator proportional gain Is.
t.

After that, proceeding to Fig. 8, Qou is calculated in the same procedure as in the case of the first invention, and the metering rack actuator (2) is calculated.
is driven by this output Qout;, and injection amount control suitable for each operating state is performed.

For example, when a potentiometer is used as the accelerator position sensor (7), the determination of the idling state in step S40 is performed by changing its output to a constant j51' (
! This can be done depending on whether the value is larger or smaller than a reference value. Other methods can also be used, such as using an accelerator idle switch that opens and closes at a specific accelerator position in response to accelerator operation.

For example, with a normal Bosch injection pump,
As the viscosity of the lubricating oil changes, the rack resistance also changes, so
- It may be difficult to control a high degree of closure using only the control procedure described above. Namely.

As the rack resistance increases, the hysteresis of the solenoid increases, the dead zone for the current rate of change becomes larger, and at the same time the speed for the current rate of change also changes, so as the temperature decreases, it becomes impossible to maintain high speed unless the gain is increased.

FIG. 11 shows an embodiment for solving this problem. That is, the correction coefficients GfjP and G according to the fuel temperature Tf
A map of f and I is stored in the rlOM (30) in advance, and the gain Ga is adjusted according to the detected fuel temperature Tf.
1r+P and Ga1nI with correction coefficients 0fLP and GfL
I): Therefore, each correction is made. The fuel temperature Tf can be considered to represent the temperature of the lubricating oil and air inside the pump, and step S41 for this correction is the seventh step.
It is inserted in the part indicated by the broken line in the figure.

With this type of uramasa, even if the viscosity of the lubricating oil changes depending on the temperature and the rack resistance fluctuates, proper control is always possible, and this correction is possible in the first and second inventions. It can be implemented in either case.

<Effects of the invention> Below 1. As is clear from the explanation of each example, 1. is detected, and the control gain used for proportional-integral control is automatically changed according to the operating state. Therefore, in a control device using a solenoid-type book actuator for metering rack FLRIs, it is possible to perform appropriate control according to various operating conditions. Accurate control is possible, and in the second invention, software Ll
- It is possible to make M and reduce costs.

[Brief explanation of drawings]

FIG. 1 is a conceptual system diagram of an embodiment of the present invention, and FIG.
)(lI) and (d) and (b) of Figure 3 are used in this Komei, ↑Explanatory diagram of operation of metering book actuator, No.
11r2I is a flowchart showing the control procedure. . (1) Fuel f') injection pump, (2) Metering rack actuator, (-1) Adjustment, ψ rack actuator position center, (6) Rotation speed Sensor, (7)...Accelerator position sensor, (11)...Fuel T''F temperature sensor, (21)...Control unit. (30)...ROM, (3:',)...・c
pu. Specialty '1'' Applicant: Yanmar Diesel Corporation Agent Agent: Patent Attorney Shino 1) Actual Figure 9 Figure 10 Figure 11

Claims (6)

[Claims] (1) A fuel injection pump in which a metering rack for adjusting the injection amount is driven by a solenoid-type rack actuator, a rack position detection means for detecting the position of the metering rack, and an actual value of the engine rotation speed. an accelerator position detection means for detecting the set value of the engine speed based on the accelerator position; and an accelerator position detecting means for detecting the set value of the engine speed based on the accelerator position; control means for calculating and outputting a control signal for driving the solenoid-type rack actuator by proportional-integral control, and the control means is configured to operate in a steady operating state in which at least a change in rotational speed is small and a change in rack position is small. An internal combustion engine characterized by having a function of determining whether a transient operating state has a large change in rotational speed or a large change in rack position, and is configured to change the control gain of proportional-integral control according to each operating state. control device. (2) The control device for an internal combustion engine according to claim 1, wherein the control means has a function of also determining the starting operation state. (3) A control device for an internal combustion engine according to claim 1 or 2, which detects the temperature of fuel supplied to the engine and corrects the control gain according to the fuel temperature. (4) A fuel injection pump in which a metering rack for adjusting the injection amount is driven by a solenoid-type rack actuator, a rack position detection means for detecting the position of the metering rack, and a means for detecting the actual value of the engine rotation speed. an accelerator position detection means for detecting the set value of the engine speed based on the accelerator position; and an accelerator position detecting means for detecting the set value of the engine speed based on the accelerator position; control means for calculating and outputting a control signal for driving the solenoid-type rack actuator by proportional-integral control, and the control means has at least a function of discriminating between a steady operating state and an idling operating state. A control device for an internal combustion engine, characterized in that a control gain of proportional-integral control is changed according to each operating state. (5) The control device for an internal combustion engine according to claim 4, wherein the control means has a function of also determining the starting operation state. (6) A control device for an internal combustion engine according to claim 4 or 5, wherein the temperature of the fuel supplied to the engine is detected and the control gain is corrected in accordance with the fuel temperature.