JP2712288B2 - Fuel injection control device for diesel engine - Google Patents

Description

Description: Object of the Invention [Industrial application field] The present invention relates to a fuel injection control device of a diesel engine, particularly a diesel engine for an automobile.

[Prior Art] Conventionally, in order to improve the fuel efficiency of a diesel engine, maintain a suitable output performance, and prevent the generation of smoke, etc., adjustment of a fuel injection pump and the like is performed at the time of shipment of a vehicle or the like. In addition to suppressing the variation in injection amount,
Various techniques have been proposed for performing precise fuel injection control in consideration of changes in the injection amount due to aging.

For example, Japanese Unexamined Patent Publication No. 56-75928 discloses that the fuel injection amount in each operating state of a diesel engine is calculated based on a calculation method of the fuel injection amount that is optimally set at the time of shipment, and the result is set in a state where the diesel engine is in an idle state. There is disclosed a technique for uniformly correcting a fuel injection amount based on a correction amount calculated in consideration of a change in fuel injection amount due to a change over time (hereinafter, referred to as an idling correction amount).

[Problems to be Solved by the Invention] According to the above-described technology, unexpected fluctuations in the rotational speed, an increase in the amount of smoke generated, and the like can be avoided by considering fluctuations in the fuel injection amount due to aging. However, the following problems remain.

In the related art, as described above, the fuel injection amount at the time of high load operation or the like is corrected and calculated by controlling the fuel injection amount in another operation state based on the correction amount at the time of idling. , Large amounts of smoke may occur. Further, a cylinder is mounted in communication with the pressurizing chamber of the fuel injection pump to perform pilot injection, and the pressure in the pressurizing chamber is increased or decreased by pressing and moving a piston sliding in the cylinder with a piezoelectric actuator. In such a fuel injection pump, for example, when switching between execution and stop of pilot injection, a large decrease in torque occurs.

As described above, the decrease in output, the generation of smoke, and the decrease in torque occur only when the fuel injection amount (hereinafter, referred to as a calculated injection amount) corrected and calculated to be optimal by the idling correction amount and the actual combustion of the diesel engine. A difference occurs between the fuel injection amount injected into the chamber (hereinafter, referred to as the actual injection amount), and the actual injection amount does not match the fuel injection amount required by the diesel engine (hereinafter, referred to as the required injection amount). Because. The cause of such a difference can be explained as follows.

Since a minute gap exists between the plunger and the cylinder of the fuel injection pump, in each seal portion such as packing, or in a seal portion in a piping route, etc., fuel leaks from the minute gap (hereinafter, the amount of fuel leakage is referred to as a leak amount). ΔQlos
s). This leakage amount ΔQloss corresponds to the difference between the calculated injection amount and the actual injection amount. Further, when the fuel injection pump is in the fuel pressure feeding stroke and the pressure in the pressurizing chamber is high, the minute space between the cylinder inner circumference and the piston outer circumference of the piezoelectric actuator,
That is, the amount of leakage of the fuel that has entered the sliding portion of the piston and returns to the pressurizing chamber also affects the difference between the calculated injection amount and the actual injection amount, equivalent to the leakage amount ΔQloss (hereinafter, including this leakage amount). The amount of leakage ΔQloss).

On the other hand, light oil, which is the fuel of a diesel engine, has a characteristic that its characteristics change with temperature, unlike gasoline. For this reason, for example, the temperature of the fuel pump is low immediately after the start of the diesel engine, etc., and when the fuel temperature is low, the kinematic viscosity is high and the fluidity is low, but for example, the temperature of the fuel pump rises with the operation of the diesel engine, and the fuel As the temperature rises, a change in fuel properties such as a decrease in kinematic viscosity and an increase in fluidity occurs. And
If the kinematic viscosity decreases and the fluidity increases due to the fuel temperature, the leakage amount ΔQloss increases, the actual injection amount decreases, and a decrease in output, a decrease in torque, and the like occur.

The ratio of the leakage amount ΔQloss to the total injection amount is not uniform depending on the operation state of the diesel engine, such as the rotation speed and the load.
At the same rotational speed, it increases as the fuel injection amount increases. That is,
The lower the rotation speed and the higher the injection amount, the more affected the fuel properties.

As described above, since the leakage amount ΔQloss is not uniform depending on the operation state of the diesel engine, the conventional technique of uniformly calculating and controlling the fuel injection amount in all operation states based on the idling-time correction amount and performing the injection control is the above-described leakage amount. Quantity ΔQl
Since the oss cannot be considered, the actual injection amount fluctuates greatly, which causes a decrease in output, generation of smoke, a decrease in torque, and the like. In particular, it is preferable to reduce the maximum injection amount, which is the upper limit value for suppressing the generation of smoke when the rotation speed becomes equal to or more than the predetermined value, but the calculated injection amount calculated by uniform correction using the idling correction amount is the maximum injection amount. The amount greatly exceeds the injection amount.
Even the actual injection amount obtained by subtracting the above may exceed the maximum injection amount. In such a case, not only a large amount of smoke is generated but also the engine is damaged.

The present invention has been made in order to solve the above problems, and an object of the present invention is to provide a fuel having a required injection amount most desirable for a diesel engine in each operation state of the diesel engine.
An object of the present invention is to provide a diesel engine fuel injection control device capable of actually performing injection control into a combustion chamber in all operating states regardless of changes in fuel properties and the like.

Configuration of the Invention [Means for Solving the Problems] A fuel injection control device for a diesel engine according to the present invention for solving the above problems includes an operating state detecting means M1 for detecting an operating state including at least the rotational speed of the diesel engine EG. And an injection amount calculation unit M2 for calculating a fuel injection amount to be injected into the diesel engine EG according to a detection result of the operation state detection unit M1, and a state of determining whether the diesel engine EG is in an idle state. When the determination means M3 and the diesel engine EG are in an idle state,
A deviation calculating means M4 for calculating a deviation ΔNe of the rotation speed of the diesel engine EG from a target idle rotation speed; and the fuel injection to reduce the deviation ΔNe to a predetermined value or less when the diesel engine EG is in an idle state. A reference correction amount calculating means M5 for calculating a reference correction amount for increasing or decreasing the amount; and an injection control means M6 for controlling the injection of fuel in an amount obtained by correcting the fuel injection amount calculated by the injection amount calculating means M2 with the reference correction amount. In the diesel engine fuel injection control device, when the diesel engine EG is not in the idle state, the fuel injection control device changes according to the rotation speed of the diesel engine EG, and becomes smaller as the rotation speed becomes higher at a predetermined rotation speed or more. It is characterized in that a leak correcting means M7 for correcting the reference correction amount based on a correction value is provided.

[Operation] In the diesel engine fuel injection control device of the present invention, the operating state detecting means M1 detects an operating state including at least the rotational speed of the diesel engine EG, and the injection amount calculating means
M2 is a detection result of the operating state detection means M1, for example, intake air temperature,
The fuel injection amount to be injected into the diesel engine EG is calculated according to the supercharging pressure, the number of revolutions, and the like. Also, the state determination means M3
It is determined whether or not the diesel engine EG is in an idle state.

The deviation calculating means M4 generates a deviation ΔNe of the rotational speed of the diesel engine EG from the target idle rotational speed when the diesel engine EG is in an idle state. This deviation Δ
Ne is caused by aging, leakage ΔQloss, adjustment variation of the injection pump, and the like.

When the diesel engine EG is in the idling state, the reference correction amount calculation means M5 increases or decreases the fuel injection amount that must be increased or decreased to set the deviation ΔNe to a predetermined value or less, for example, 0, that is, sets the reference correction amount to a predetermined value. Or a predetermined map or the like.

The injection control means M6 controls the injection of fuel in an amount obtained by correcting the fuel injection amount calculated by the injection amount calculation means M2 with the reference correction amount.

Therefore, when the diesel engine EG is in the idle state, the amount of fuel that is obtained by correcting the fuel injection amount with the reference correction amount is injected, so that the idling time due to aging, leakage amount ΔQloss, injection pump adjustment variation, etc. It is possible to avoid unexpected fluctuations in the number of revolutions and an increase in the amount of smoke generated.

On the other hand, when the diesel engine EG is not in the idling state, the leakage correction means M7 performs a reference correction based on a leakage correction value that changes according to the rotation speed of the diesel engine EG and becomes smaller as the rotation speed becomes higher above a predetermined rotation speed. Correct the amount. The leak correction value can be calculated by a calculation formula or obtained from a map, for example.

This leak correction value is for correcting the leak amount ΔQloss, which is not uniform depending on the operating state of the diesel engine. As described above, the leakage amount ΔQlos
s increases as the fuel injection time increases, that is, as the fuel injection amount increases, as the rotational speed decreases, and as the fuel injection amount increases at the same rotational speed, the s increases. Therefore, normally, the leakage amount ΔQloss increases with an increase in the rotational speed and then decreases with an increase in the rotational speed. The leak correction value corresponds to such a change in the leak amount ΔQloss.

Then, the injection control means M6 controls the fuel injection amount calculated by the injection amount calculation means M2 with a reference correction amount corrected based on the leak correction amount, thereby controlling the fuel injection. The injection amount when not in the idling state is a leakage amount ΔQl that fluctuates depending on the number of revolutions, load, and the like.
oss is considered, and the optimal fuel injection amount at that time is obtained. That is, since there is no difference between the required injection amount and the actual injection amount, it is possible to prevent problems such as insufficient output and smoke caused by the difference.

Next, an embodiment of a diesel engine fuel injection control device according to the present invention will be described with reference to the drawings.

FIG. 2 is a system configuration diagram of a diesel engine according to a first embodiment of the present invention.

The diesel engine distribution type fuel injection pump 1 is driven by rotation of a drive pulley 3 connected to a crankshaft of the diesel engine 2 via a belt or the like, and pumps fuel to a fuel injection nozzle 4 of the diesel engine 2. The drive pulley 3 has a projection 5 projecting therefrom. The reference cam angle sensor 7 provided on the pump housing 6 of the fuel injection pump 1 is used to determine a predetermined crank angle of the diesel engine 2 (in this embodiment, TDC (top dead center). Point)) can be detected. A drive shaft 8 of the fuel injection pump 1 connected to the drive pulley 3 is provided with a vane pump 9 as a fuel feed pump and a pulsar 10 having a plurality of projections on its outer peripheral surface. It is connected to the cam plate 11 via a coupling.

The cam plate 11 is integrally joined with the plunger 12,
It is rotated according to the rotation of the drive shaft 8. Further, the cam plate 11 is connected to a roller ring 14 positioned by a timer device 13, and is reciprocated in the left-right direction in the figure by a cam roller 15 attached to the roller ring 14. Therefore, the cam plate 11 and the plunger 12 are rotated and reciprocated by the rotation of the drive shaft 8.

Next, the plunger 12 is connected to a fuel cut valve (FC) (not shown).
U) is inserted into a pump cylinder 17 which is communicated with a fuel chamber 16 in the pump housing 6 via an intake port opened and closed by U), pressurizes the fuel by its reciprocating motion, and delivers the diesel engine 2 via a delivery valve 18. Pumps fuel to each cylinder. That is, a fuel passage 12a corresponding to the number of cylinders is formed at the tip of the plunger 12, and when moving in the left direction in the drawing, the fuel in the fuel chamber 16 is sucked into the pressurizing chamber 17a, and the fuel passage 12a moves rightward in the drawing. When moving, the fuel in the pressurizing chamber 17a is pressurized and the fuel is pumped from the distribution port 12b.

On the other hand, a spill port 17b is formed from the pump cylinder 17 to the housing 6 and communicates with the pressurizing chamber 17a of the cylinder 17, and is communicated with the fuel chamber 16 via the electromagnetic spill valve 20. The electromagnetic spill valve 20 is operated by opening and closing the needle valve 20a, and when the plunger 12 moves to the right in the drawing, that is, at the time of fuel pressurization and pressure feeding, the pressurizing chamber 17a and the fuel chamber 16 are communicated.
The fuel in 7a overflows and the fuel supply to the diesel engine 2 is stopped. The fuel passage 12a of the plunger 12 communicates with the fuel introduction passage 17c of the cylinder 17, which is opened by the fuel cutoff valve 21 in the suction stroke and is shut off in other strokes.

Next, the timer device 13 includes a timer housing 13a, a timer piston 13b fitted into the timer housing 13a, and connected to the roller ring 14, and a spring 13c for urging the timer piston 13b rightward in the drawing. By positioning the timer piston 13b by the fuel pressure of the high pressure chamber 13d into which the high pressure fuel in the fuel chamber 16 is introduced,
The position of the roller ring 14 is determined, and the fuel injection timing is adjusted. The fuel pressure in the high-pressure chamber 13d is equal to the high-pressure chamber 13d and the low-pressure chamber 13e.
The pressure is regulated by a hydraulic control valve 23 which is provided in a communication passage 22 with which the opening and closing are controlled by a pulse drive signal having a controlled duty ratio.

A rotation speed sensor that generates a detection signal every time a protrusion formed on the outer peripheral surface of the pulsar 10 crosses the roller ring 14 positioned by the timer device 13 and the hydraulic control valve 23 at a position facing the pulsar 10. And an actual cam angle sensor (hereinafter also referred to as a rotational speed sensor) 25 which is also capable of detecting the rotation speed of the fuel injection pump, that is, the engine rotation speed of the diesel engine 2 and the fuel injection cycle of the fuel injection pump. Have been. That is, since 56 projections are formed on the outer peripheral surface of the pulsar 10 by cutting the outer peripheral surface into four equal parts at four points, the detection signal from the actual cam angle sensor 25 is shaped into a waveform. A reference signal synchronized with the fuel injection cycle and a reference cam angle signal representing the rotation speed are obtained. Since the actual cam angle sensor 25 is fixed to the roller ring 14 and moves with its rotation, the actual cam angle sensor 25 can detect the lift of the cam roller 15, that is, the fuel injection start timing, from the reference signal and the actual cam angle signal. The TDC signal of the diesel engine 2 can be obtained by shaping the waveform of the detection signal from the reference cam angle sensor 7 described above.

In the diesel engine 2, a main combustion chamber 35 is formed by a cylinder 33 and a piston 34, and a sub-combustion chamber 36 having a glow plug 36 a is connected to the main combustion chamber 35. Fuel is injected into the sub-combustion chamber 36. A turbocharger 38 is provided in the intake pipe 37 of the diesel engine 2.
The compressor 39 is disposed, while the exhaust pipe 40 is provided with a turbine 41. The exhaust pipe 40 is also provided with a waste gate valve 42 for adjusting the supercharging pressure.

As the detector, the rotational speed sensor 25 of the fuel injection pump 1, the accelerator sensor 51 including a potentiometer for detecting the accelerator operation amount, and the intake air temperature for detecting the intake air temperature, which are provided in the intake pipe 37 of the diesel engine 2, are provided. A sensor 52 is provided at an intake port 37a communicating with the intake pipe 37, and a supercharging pressure sensor 53 for detecting a supercharging pressure, and a water temperature sensor 5 provided for the cylinder block 33a for detecting a cooling water temperature.
4, an air conditioner switch 55 for instructing driving of the compressor of the air conditioner 55a, a power steering switch 56 for indicating that the power steering is operating, a neutral switch 57 for indicating that the shift of the automatic transmission is neutral, and an axle. A vehicle speed sensor 58 that detects the speed of the vehicle by replacing the NS pole of the rotating magnet with an on / off signal of a reed switch is provided.

The detection signals of each of the above sensors are sent to an electronic control unit (hereinafter simply referred to as EC
The ECU 60 drives the fuel cutoff valve 21, the electromagnetic spill valve 20, and the hydraulic control valve 23 to control the diesel engine 2, and further controls the glow plug 36a and the air conditioner. The on / off control of the electromagnetic clutch 59 for transmitting the driving force from the diesel engine to the 55a compressor is performed.

 Next, the configuration of the ECU 60 will be described with reference to FIG.

The ECU 60 is a central processing unit that inputs and calculates each signal detected by each of the above-described sensors according to a control program, and performs processing for controlling each of the valves 20, 21, and 23, the glow plug 36a, and the electromagnetic clutch 59. (Hereinafter simply referred to as a CPU) 60a, a read-only memory (hereinafter simply referred to as a ROM) 60b in which the control program and initial data are stored in advance, and various data input to the ECU 60 and data necessary for arithmetic control are temporarily stored. Random access memory (hereinafter simply referred to as RAM) 60c and various data necessary for controlling the diesel engine 2 can be stored and held even if a key switch (not shown) of the diesel engine 2 is turned off by a driver. The backup random access memory backed up by the battery The referred to as the backup RAM) 60d or the like around the is configured as a logic operation circuit performs input and output of the external each device is connected to the input port 60f and the output port 60g through a common bus 60e.

Also, the ECU 60 includes buffers 60h, 60i, and 60j for output signals from the accelerator sensor 51, the water temperature sensor 54, the intake air temperature sensor 52, the supercharging pressure sensor 53, the air conditioner switch 55, the power steering switch 56, and the neutral switch 57. , 60k, 60
m, 60n, 60p are provided, and some of the above sensors 51, 52, 5
A multiplexer 60q for selectively outputting the output signals of the 3,54 to the CPU 60a, an A / D converter 60r for converting an analog signal to a digital signal, an output signal of the rotation speed sensor 25, a reference cam angle sensor 7, and a vehicle speed sensor 58. A waveform shaping circuit 60s for shaping the waveform is also provided. Signals from these sensors are input to the CPU 60a via the input port 60f.

Further, the ECU 60 includes the electromagnetic spill valve 20, the fuel cutoff valve 21, the hydraulic control valve 23, the electromagnetic clutch 59, the glow plug
36a drive circuits 60t, 60u, 60v, 60w, 60x, and the CPU 60a drives the above drive circuits 60t, 60u, 60v, 60w, 6 via the output port 60g.
Output control signal to 0x.

Next, the processing executed by the ECU 60 will be described based on the flowchart shown in FIG.

First, the flowchart of FIG.
FIG. 5 shows an interrupt routine for calculating the number of revolutions of the engine. An interrupt request signal is generated by a pulse signal from a rotation speed sensor 25 attached to the injection pump 1, and the time between pulses shown in the graph of FIG. (I) (step 100), the number of data for one revolution of the engine,
Cyclic accumulation in memory (Step 11
0). Thereby, data ΔT (i) for one rotation from the most recent rotation data is accumulated, and data past one rotation or more is deleted.

The flowchart of FIG. 6 shows an interrupt routine for calculating the vehicle speed, in which an interrupt request signal is generated based on the detection signal of the vehicle speed sensor 58, and the time between pulses is set to a variable TSP.
D (step 120), and the vehicle speed SPD is obtained from the variable TSPD.
= K1 / TSPD (K1: constant) is calculated (step 130).

The flowcharts of FIGS. 7A and 7B show an injection amount calculation routine in which a key switch (not shown) is repeatedly processed from the time when it is turned on until it is turned off. (Step 20
0). Next, the rotation speed Ne = K2 / ΔT (i) (K2: constant) is calculated from the interpulse time data ΔT (i) for one rotation of the diesel engine 2 accumulated in the rotation speed interruption routine of FIG. This value is set as the current engine speed (step 210), and the accelerator opening Acc is calculated from the output value from the accelerator sensor 51 (step 220).

Thereafter, a target idle speed (NIDL) is calculated according to the engine cooling water temperature THW, the ON / OFF state of the air conditioner, and the operation state determined by the selection state of the D lens or the N range (step 300). The flowchart of FIG. 8 shows the target idle speed calculation logic in step 300 described above. First, the engine cooling water temperature THW is calculated from the water temperature sensor 54 (step 310), and a water temperature correction coefficient F (w) corresponding to the cooling water temperature THW is calculated using, for example, a characteristic as shown in the graph of FIG. Step 320). Next, in the case of a torque converter vehicle, the neutral switch 57 determines whether the range is the neutral (N) range or the drive (D) range (step 330). If in D range, go to step 340
In the case of ranges, the process jumps to step 345, and multiplies the target idle speeds ND (D range) and NN (N range) of each range by the water temperature correction coefficient F (w) obtained in step 320.
ND and NN are target rotation speeds in a completely warmed-up state. When the cooling water temperature is low, the target rotation speed is increased by a multiple of F (w) to achieve warm-up idle-up. The target idle speed corresponding to the cooling water temperature THW in each range of the torque converter thus determined is NF. In steps 350 and 355, the ON / OFF of the air conditioner switch 55 in each range is determined. In steps 360 and 365, when the air conditioner switch 55 is ON, the idle-up rotation speed NDAC (D range) and NNAC (N range) are set to the target idle speed described above. Add to the number NF. The target idle speed NF corresponding to the load state thus obtained is set to NIDL (step 370). If there is an idle-up request signal from the power steering switch 560N, the target idle rotation speed NF is added by the corresponding idle-up request rotation speed.

Returning to the injection amount calculation routine of FIG.
A governor pattern movement proportional correction amount (proportional portion) NP is calculated which necessarily moves the governor pattern in accordance with the load fluctuation of the THW, neutral range, air conditioner 55a, etc. (step 400). The flowchart of FIG. 10 shows the calculation logic of the expected governor pattern movement proportional correction amount (proportional amount) NP in step 400 described above. First, the cooling water temperature T
The correction amount NPW corresponding to the HW is calculated using, for example, a characteristic as shown in the graph of FIG. 11 (step 410). Next, the output content of the neutral switch 57 is determined (step 42).
0), the process jumps to step 430 in the case of the D range, and to step 435 in the case of the N range, and determines ON / OFF of the air conditioner switch 55 in each range. And steps 440,450,46
At 0, the range of the torque converter changes, or the air conditioner switch 5
Governor pattern movement correction amount constant KN PD (D range, air conditioner 55a OFF), KN PDAC in anticipation of load fluctuation accompanying 5
(D range, air conditioner 55a ON), KN PNAC (N range,
The air conditioner 55a ON) is set to Np. Then, in step 470, the cooling water temperature correction amount Npw obtained in step 410 is added to Np to obtain the final expected governor pattern movement proportional correction amount Np. If the power steering 56 is ON, the estimated correction amount is added to Np.

Returning to FIG. 7 (a), it is determined whether the operation state of the diesel engine 2 is any of the idle stable state, the idle unstable state, and the running state by the determination from step 500 to step 540. That is, it is determined whether or not the vehicle speed SPD obtained by the vehicle speed sensor 58 is zero (step 50).
0), if SPD = 0, it is determined from the accelerator opening ACC determined in step 220 whether or not the engine is in an idle state (step 510). If the engine is idling, it is determined whether or not the engine has just started after determining whether the rotation speed Ne obtained in step 210 satisfies Ne> 400 rpm (step 520), and Ne> 400.
If the rpm is satisfied, the process proceeds to step 530. Step 530
Then, the elapsed time from when all the conditions of 500, 510 and 520 are satisfied is counted by a counter CTIME of, for example, 5 msec (or 5 to 50 msec). Then, it is determined whether 1.5 seconds or more have elapsed (step 540). If it is determined in step 540 that the elapsed time has exceeded the predetermined time (1.5 seconds), it is determined that the vehicle is in the idle stable state, the idle flag FIDL indicating the state is set (step 550), and then the process proceeds to step 560.

On the other hand, if it is determined in step 500 that the vehicle speed SPD> 0, or if it is determined in step 510 that the vehicle is not in the idle state, it is determined that the vehicle is in the traveling state, and the process proceeds to step 740 and subsequent steps described later. Further, at step 520
If it is determined that Ne> 400 rpm is not established, CTIME is cleared (step 570), and for example, it is determined that the cooling water temperature THW is low, and that the engine is in an unstable idle unstable state immediately after starting. After that, the idle flag FIDL is reset (step 580), including the case where it is determined in step 540 that 1.5 or more has not elapsed (step 580), and the process proceeds to step 740 described later.

After setting FIDL in step 550, an error ΔNIDL between the target idle speed NIDL obtained in step 300 and the current actual speed Ne obtained in step 210 is calculated (step 560). Here, the target idle speed DIDL in step 300 may be calculated. Next, the governor pattern movement integral correction amount (integral amount) NI is calculated based on the error based on ΔNIDL (step 600). The flowchart of FIG. 12 shows the calculation logic of the governor pattern movement integral correction amount NI in step 600 described above. First, the correction integration amount ΔNI is set to Δ
From the NIDL, a predetermined relationship, for example, a characteristic as shown in the graph of FIG. 13 is obtained by a calculation formula or map interpolation (step 610). Next, the correction integration amount ΔNI obtained in step 610
Is added and integrated to obtain ΣΔNI (step 620), and step 6
It is determined whether or not the integrated value ΣΔNI at 20 has deviated from the upper and lower limits (here, both upper and lower limits are KN IMAX as absolute values) (step 630).
Use to guard the upper limit (+ KN IMAX) and lower limit (-KN IMAX). The governor pattern movement integral correction amount ΣΔNI thus obtained is set as NI (step 650).

Returning to the injection amount calculation routine of FIG. 7 (b), the sum of the governor pattern proportional correction amount NP obtained in step 470 and the integral correction amount NI obtained in step 650 is set as NPI (step 710). Next, the integral correction amount NI set in step 650 is averaged by a smoothing process as shown in the following equation to obtain a learning value NFG (step 720).

 NFG = (ΣNI + NI) / n where n is the number of times the integral correction amount NI is calculated.

If the learning value NFG is obtained by setting the value of n to be as large as possible in step 720, the learning speed of the NFG is increased, and an erroneous learning value NFG due to the influence of various disturbance factors may be calculated. Can be avoided. In addition, the possibility of erroneous learning of the NFG may be reduced in consideration of the offset based on the ON / OFF condition of the air conditioner with respect to the integral correction amount NI.

Next, the basic correction amount Qco, which is the injection correction amount in the idling stable state, is calculated by the following formula by a map search or a calculation formula according to the learning value NFG calculated in step 720 (step 730).

Qco = KQ × NFG (KQ: constant) The calculated basic correction amount Qco is the backup RAM
The previous Qco is updated and written to the predetermined address of 60d,
It is stored.

Next, the NPI obtained in step 710 is replaced with the actual rotational speed Ne.
Then, the rotation speed Neo for calculating the injection amount is calculated by subtracting the following expression from the following expression (step 740).

Neo = Ne-NPI Based on the Neo and the accelerator opening Acc calculated in step 220, the basic injection amount QBASE is obtained by a map search or a calculation formula (step 800), thereby apparently changing the governor pattern to the rotation speed axis. In the direction by NPI.
FIG. 14 is a graph showing this state, and shows a governor pattern II in which the idle injection amount governor pattern I is translated in the direction of the rotational speed axis by NPI. The maximum injection amount (smoke limit) QMAX according to the operation state at this time based on the intake air temperature, the supercharging pressure, the rotation speed, and the like.
(Step 810).

Next, the idle flag F indicating an idle stable state
It is determined whether the IDL is in the set state (step 820). F
If the IDL is in the set state, it is determined that the vehicle is in the idling stable state or the vehicle is in the running state after the state, and the final correction amount Qc is calculated as follows (step 830). That is, from the basic correction amount Qco calculated in step 730 and stored in the backup RAM 60d and the coefficient f (Ne) as shown in FIG. Calculate the final correction amount Qc,
The calculated Qc is updated to a predetermined address of the backup RAM 60d by updating the previous Qc and stored (step 830).

Qc = Qco × f (Ne) Here, the above-mentioned coefficient f (Ne) changes in fuel properties (kinematic viscosity and fluidity) with refueling and a change in fuel temperature, and leaks from a minute gap such as a fuel injection pump. Leakage amount ΔQloss
Is assumed and is defined as follows.

As described above, the leakage amount ΔQloss increases as the fuel injection time is longer, that is, as the rotation speed is lower for the same fuel injection amount, and is increased as the fuel injection amount is larger at the same rotation speed. Accordingly, as shown in FIG. 16, the lower the rotation speed and the higher the injection amount condition, the more affected by the leakage amount ΔQloss. Qi in the figure
Indicates the fuel injection amount, and indicates that the injection amount decreases in the order of Qi, Qi + 1, Qi + 2,. On the other hand, the maximum injection amount QMAX obtained in step 810 generally increases as the rotational speed increases, as shown in FIG. 17, then becomes constant in a range of a predetermined rotational speed or more, and decreases as the rotational speed further increases. It is determined to be. Therefore, the leakage amount ΔQlos
The coefficient f (Ne) as shown in FIG. 15 assuming s increases from the above-described FIGS. 16 and 17 with respect to the rotation speed Ne as the rotation speed increases, and then decreases as the rotation speed increases. They have a relationship.

On the other hand, if the idle flag FIDL is in the reset state in step 820, the current state is the unstable idle unstable state immediately after the start, or the running state has passed through this state, and the NFG and Qco have been calculated. It is determined that there is not. Therefore, the previous final correction amount Qc stored and held in the backup RAM 60d is read (step 840), and then the present final correction amount Qc is calculated from the read previous Qc by the following equation. Backup RAM
The data is stored in 60d (step 850).

Qc = Qc × k Here, k is a constant satisfying 0 ≦ k ≦ 1, and is set by the engine coolant temperature THW or the like.

When the final correction amount Qc is obtained in step 830 or 850 in this manner, the maximum injection amount QMAX calculated in step 810
To the current maximum injection amount QMAX
(Step 860).

QMAX = QMAX + Qc Next, the smaller value of QBASE obtained in step 800 and QMAX obtained in step 860 is set as a final injection amount QFIN (step 870), and an injection amount command value Vs corresponding to the obtained final injection amount QFIN is obtained. Output to the injection amount control actuator drive circuit (step 880).

As described above, the present embodiment calculates the basic correction amount Qco based on the learning value NFG obtained by smoothing and averaging the governor pattern movement integral correction amount NI in the idle stable state, and the Qco, The maximum injection amount QMAX in the running state is corrected at any time by the final correction amount Qc calculated from the coefficient f (Ne) determined by the rotation speed assuming the leakage amount ΔQloss. Therefore, the fuel injection control device according to the present embodiment appropriately sets the maximum injection amount QMAX according to the rotation speed as described above regardless of a change in the fuel property, that is, when the leakage amount ΔQloss is small, QMAX is small and ΔQloss is large. Sometimes Q
By greatly correcting MAX, the fuel injection control device can prevent the actual injection amount from becoming too large, reliably avoid the occurrence of a large amount of smoke and damage the engine, and maintain stable output performance.

Further, when the fuel injection pump is mounted on a diesel engine, the fuel injection amount variation caused by the so-called in-cylinder-internal difference and the fuel injection amount variation caused by the adjustment variation of each of the above-described injection pumps are caused by the above-mentioned leakage amount ΔQloss, This is reflected in the difference between the amount and the actual injection amount. For this reason, the fuel injection amount variation can be corrected.

In particular, the fuel injection control device according to the present embodiment can perform fuel amount correction based on a change in fuel properties without using a special sensor such as a fuel temperature sensor, so that the fuel injection control device is inexpensive.

Further, in an unstable idle unstable state immediately after the start, or in a traveling state after the state and before the idle stable state, the present final correction amount Qc is changed to the previous final correction amount Qc.
And a constant k (0 ≦ 0) determined by the engine cooling water temperature THW, etc.
k ≦ 1). For this reason, the final correction amount Qc is not set to an unnecessarily large value, and the generation of smoke and the like can be suppressed even in the idle unstable state or the like, and unnecessary load is not applied to the engine.

Next, a second embodiment of the present invention will be described. The second embodiment has the same system configuration as the first embodiment described above, and executes an injection amount calculation routine as shown in the flowcharts of FIGS. 18 (a) and 18 (b).

The flowcharts in FIGS. 18 (a) and 18 (b) are similar to those in the first embodiment, except that a key switch (not shown) is turned on and then turned off.
9 shows an injection amount calculation routine that is repeatedly performed until the injection amount is calculated.

As in the first embodiment, the processing from the initial processing in step 1200 to the processing for calculating the maximum injection amount QMAX in step 1810 is performed. That is, first, an initial process for resetting a flag or the like is performed (step 1200), and the current rotation speed Ne (step 121) is performed.
0), the accelerator opening Acc (step 1220) is calculated. Next, a target idle speed (NIDL) according to the operating state is calculated (step 1300), and an expected governor pattern movement proportional correction amount NP is calculated from the engine coolant temperature THW and the like (step 1400). Next, it is determined whether or not the engine is in the idling stable state based on the determinations from step 1500 to step 1540. If the idle state is stable, the idle flag FIDL is set (step 1550), and an error ΔNIDL between the target idle speed NIDL and the current speed Ne is calculated (step 1560).

On the other hand, if a negative determination is made in step 1500 or 1510, the flow shifts to step 1740. If a negative determination is made in step 1520, CTIME is cleared (step 157).
0), the idle flag FIDL is reset including the case where the determination is made in step 1540 (step 1580), and the process proceeds to step 1740.

The governor pattern movement integral correction amount NI is calculated based on the error ΔNIDL calculated in step 1560 (step 160).
0). Thereafter, the sum of the governor pattern proportional correction amount NP and the integral correction amount NI is set as NPI (step 1710). Next, the integral correction amount NI is smoothed and averaged to obtain a learning value NFG.
Is calculated (step 1720), the basic correction amount Qco is calculated by a map search or a calculation formula according to the NFG, and the basic correction amount Qco is stored and held in the backup RAM 60d (step 1730).

Next, the NPI obtained in step 1710 is subtracted from the actual rotation speed Ne to calculate an injection amount calculation rotation speed Neo (step 1740), and the basic injection amount Q is calculated from the Neo and the accelerator opening Acc.
BASE is calculated (step 1800). Next, the maximum injection quantity Q
MAX is determined (step 1810), and the smaller value of QBASE and QMAX is set as the final injection amount QFIN (step 1900).

Then, it is determined whether or not the idle flag FIDL indicating the idle stable state is in the set state (step 191).
0). If the FIDL is in the set state, it is determined that the vehicle is in the idling stable state or the vehicle is in the running state after the state, and the final correction amount Qc is calculated as follows (step 1920). That is, based on the basic correction amount Qco calculated in step 1730 and stored in the backup RAM 60d and the coefficient f determined from a map as shown in FIG.
(Ne, L) and the final correction amount Qc is calculated as in the following equation, and the calculated Qc is stored in a predetermined address of the backup RAM 60d.
Is updated and stored (step 1920).

Qc = Qco × f (Ne, L) It should be noted that f (Ne, L) is determined from the rotation speed Ne or the load L by search or map interpolation according to both of them.

Here, the coefficient f (Ne, L) is the amount of leakage ΔQlos that changes in fuel properties (kinematic viscosity and fluidity) with changes in fuel supply and fuel temperature and leaks from a minute gap such as a fuel injection pump.
s is assumed and is defined as follows. That is, at the same load, the smaller the rotation speed, the smaller the value. F (Ne1, Li)> f (Nei, Li)> f (Nen, Li) At the same rotation speed, the larger the load, the larger the f (Ne1, L1). <F (Nei, Li) is a two-dimensional map of <f (Nei, Lm) relation.

On the other hand, if the idle flag FIDL is reset in step 1910, NFG and Qco are calculated because the vehicle is currently in an unstable idle unstable state immediately after starting or is in a running state through this state. It is determined that there is not. Therefore, the previous final correction amount Qc stored and held in the backup RAM 60d is read (step 1930). Next, the last correction amount Qc of this time is calculated from the previous Qc read by the following equation, and the calculated Qc is used as a backup RA in the same manner as in step 1920.
It is stored in the M60d (step 1940).

Qc = Qc × k Here, k is a constant satisfying 0 ≦ k ≦ 1, and is set by the engine coolant temperature THW or the like.

In this way, in step 1920 or 1940, the final correction amount Qc
Is obtained, the final injection amount Q calculated in step 1900
FIN is added with the final correction amount Qc to obtain the final injection amount QF
Find IN (step 1950). Instead of adding Qc, the final injection amount QFIN may be calculated by multiplying QFIN by a coefficient determined according to Qc.

QFIN = QFIN + Qc Next, an injection amount command value VS corresponding to the obtained final injection amount QFIN is obtained and output to the injection amount control actuator drive circuit (step 1960).

As described above, in the second embodiment, the governor pattern movement integral correction amount NI
The basic correction amount Qco is calculated based on the learning value NFG obtained by the smoothing and averaging process, and the leakage amount ΔQlos in an operation state determined by the Qco and the rotation speed or the load or both.
Final correction amount Q calculated from coefficient f (Ne, L) assuming s
With c, the final injection amount QFIN in the running state is corrected as needed. Therefore, according to the fuel injection control device of the second embodiment, the same effects as those of the first embodiment are apparent. That is, the injection amount most desirable for the diesel engine can be actually controlled to be injected into the combustion chamber irrespective of the change in the fuel property. For this reason, the same effects as those of the first embodiment can be obtained, such as a reduction in output and a large amount of smoke generated, and a stable output performance can always be maintained.

In particular, the fuel injection amount control device of the second embodiment can calculate the final injection amount QFIN as the optimum fuel injection amount for the diesel engine based on the two state quantities (rotational speed and load) that determine the operating state. A fuel injection device that can perform more precise and precise fuel injection control, and can always and reliably prevent output reduction and smoke generation even during braking operation, and maintain stable output performance Becomes

 Next, a third embodiment of the present invention will be described.

As shown in FIG. 20, in the third embodiment, the piezoelectric actuator 100 is installed in the distribution type fuel injection pump 1 for a diesel engine of the first and second embodiments.
Therefore, common constituent members are denoted by the same reference numerals, description thereof will be omitted, and a configuration different from the above-described embodiment will be described.

That is, the pressurizing chamber of the pump cylinder 17 of the fuel injection pump 1
A piezoelectric actuator 100 is provided in communication with 17a. This piezoelectric actuator 100 has a cylindrical casing 1
01 forms a body, one end of which is threaded 10
It is fixed to the housing 6 of the fuel injection pump 1 via 2. The casing 101 serves as a cylinder, and includes therein a piezoelectric element 103 and a piston 104 which is in contact with the piezoelectric element 103 and is slidable inside the casing 101 while maintaining liquid honey. ing. The piezoelectric element 103 is configured by laminating a plurality of disk-shaped PZTs, and expands and contracts in the left-right direction in FIG. The right end of the piezoelectric element 103 is the casing 1
01 is fixed to the inner end face.

The piston 104 forms each knitting volume chamber 107 by the distance piece 106 fitted to the left end of the casing 101 and the inner surface of the casing 101. O-ring 108 on piston 104
Are provided so that the liquid tightness is further maintained. A disc spring 109 is provided in the variable volume chamber 107, and an initial load is applied to the piezoelectric element 103 via the piston 104. Further, a communication hole 110 is provided in the distance piece 106, and the variable volume chamber 107 and the above-described pressurizing chamber 17a of the fuel injection pump 1 communicate with each other through the communication hole 110. Also, stamp stamp piece
A minute projection 106a protruding from an end face of the pressurizing chamber 17a on the side of the pressurizing chamber 17a contacts the end face of the pump cylinder 17, and a minute space communicating with the pressurizing chamber 17a is formed on the end face. A leak passage 111 communicating this minute space with the inner surface of the casing 101 on the right side of the piston 104 is bored in the peripheral wall of the casing 101. Through the leak passage 111, the casing 101
The fuel flowing into the minute gap in the piston sliding portion on the inner surface is returned to the pressurizing chamber 17 when the plunger 12 moves rightward in the drawing, that is, when the fuel is pressurized and fed.

Note that a signal line 105 for transmitting a drive signal to the piezoelectric element 103 is provided.
Is connected to the ECU60.

The ECU 60 includes a drive circuit (not shown) connected to the piezoelectric actuator 100 and the output port 60g to drive the piezoelectric actuator 100, and drives the piezoelectric actuator 100 based on the detection results from the reference cam angle sensor 7 and the rotation speed sensor 25. A control signal indicating a timing is output to the drive circuit. The drive circuit receives the drive circuit control signal and applies a voltage corresponding to the control signal to the piezoelectric actuator 100 via the signal line 105. Thus, the piezoelectric actuator 100
The piezoelectric element 103 extends leftward in the figure to reduce the internal volume of the variable volume chamber 107.

Next, an injection amount calculation routine executed in the third embodiment of the present invention will be described with reference to the flowcharts shown in FIGS. 21 (a) and 21 (b). In the injection amount calculation routine of the present embodiment, the same processing as that of the injection amount calculation routine of the first and second embodiments described above will be briefly described.

The flowcharts of FIGS. 21 (a) and (b)
After the key switch (not shown) is turned on as in the embodiment of FIG.
The process is repeated until it is turned off, and first, an initial process of resetting a flag or the like is performed (step 2200). Next, the current rotational speed Ne (step 2210) and the accelerator opening ACC (step 2220) are calculated. Next, a target idle speed (NIDL) according to the operating state is calculated (step 2300),
The expected governor pattern movement proportional correction amount NP is calculated based on the engine cooling water temperature THW and the like (step 2400). Next, it is determined whether or not the engine is in the idling stable state based on the determinations from step 2500 to step 2540. If the idle state is stable, the idle flag FIDL is set (step 2550), and an error ΔNIDL between the target idle speed NIDL and the current speed Ne is calculated (step 256).
0).

On the other hand, if a negative determination is made in step 2500 or 2510, the flow shifts to step 2740. If a negative determination is made in step 2520, the CTIME is cleared (step 257).
0), the idle flag FIDL is reset including the case determined in step 2540 (step 2580), and the process proceeds to step 2740.

The governor pattern movement integral correction amount NI is calculated based on the error ΔNIDL calculated in step 2560 (step 260).
0). Thereafter, the sum of the governor pattern proportional correction amount NP and the integral correction amount NI is set as NPI (step 2710). Next, the integral correction amount NI is smoothed and averaged to obtain a learning value NFG.
Is calculated (step 2720), and the basic correction amount QPI0 at the time of executing the pilot injection in the idling stable state is calculated by the following formula using a map search or a calculation formula according to the NFG, and the basic correction amount QPI0 is stored in the backup RAM 60d. The information is stored (step 2730).

QPI0 = KQP × NFG (KQP: constant) Alternatively, as shown in FIG. 23, QPI0 corresponding to the NFG obtained from the predetermined relationship between NFG and QPI0 is calculated.

Next, the NPI obtained in step 2710 is subtracted from the actual rotation speed Ne to calculate an injection amount calculation rotation speed Neo (step 2740), and the basic injection amount Q is calculated from the Neo and the accelerator opening Acc.
BASE is calculated (step 2800).

Next, it is determined whether or not the current operating state of the diesel engine 2 is in a pilot injection execution region where the pilot injection by the piezoelectric actuator 100 is executed (step 3000). For example, the QBASE calculated in step 2800 above
Is less than or equal to the predetermined set value QBASE1, and
The current rotational speed Ne calculated in 210 is a predetermined set rotational speed N
The determination is made based on whether or not the engine is in a light load / low rotation range that is equal to or less than e1.

If it is determined in step 3000 that the pilot injection is in the pilot injection execution region, the pilot injection is not performed this time, so the final pilot increase QPI is set to 0 (step 3000).
3010) Go to step 3080 without performing any processing. On the other hand, if it is determined to be in the pilot injection execution region, the basic pilot QPI1 is calculated as follows (step 3020). That is, a difference Δθ between the spill angles obtained by performing an experiment for injecting the same injection amount Q1 between when the pilot injection is executed and when the pilot injection is not executed when the rotation speed Ne is the same (in this embodiment, the electromagnetic spill valve 20 is QPI1 is determined based on the time difference before driving (see FIG. 22).
When determining the QPI1, so-called map interpolation may be used based on a map in which the influence of the number of rotations, load, water temperature and the like is taken into account.

After calculating QPI1, it is determined whether or not the idle flag FIDL indicating the idle stable state is in the set state (step 3030). If the FIDL is in the set state, it is determined that the vehicle is in the idling stable state or the vehicle is in the running state after the state, and the pilot increase correction amount QPI2 is calculated as follows (step 3030). That is, the following equation is obtained from the basic correction amount QPI0 stored in the backup RAM 60d after being calculated in the step 2730 and a coefficient f (Ne) as shown in FIG. 15 previously stored in the ROM 60b based on the rotation speed Ne. The pilot increase correction amount QPI2 is calculated as described above, and the calculated QPI2 is updated to a predetermined address of the backup RAM 60d by updating the previous QPI2 and written and stored (step 304).
0).

QPI2 = QPI0 × f (Ne) Here, the coefficient f (Ne) is such that the fuel properties (kinematic viscosity and fluidity) change with the change of fuel supply and fuel temperature, and the contraction of the piezoelectric element 103 of the piezoelectric actuator 100 By piston
A leak passage 111 from the minute gap in the sliding portion 104 to the pressurizing chamber 17
Is assumed to be the leakage amount ΔQloss returned via the.

On the other hand, if the idle flag FIDL is reset in step 3030, NFG and QPI0 are calculated because the vehicle is currently in an unstable unstable unstable state immediately after starting, or is in a running state through this state. It is determined that there is not. Therefore, the previous pilot increase correction amount QPI2 stored and held in the backup RAM 60d is read (step 3050), and the present pilot increase correction amount QPI2 is calculated from the read previous QPI2 by the following equation, and step 3
The QPI2 calculated similarly to 040 is stored and held in the backup RAM 60d (step 3060).

QPI2 = QPI2 × k Here, k is a constant satisfying 0 ≦ k ≦ 1, and is set by the engine cooling water temperature THW or the like.

When the pilot increase correction amount QPI2 is obtained in step 3040 or 3060 in this way, the pilot increase correction amount QPI2 is added to the basic pilot increase QPI1 calculated in step 3020 to obtain the current final pilot increase QPI (step 3070). ).

QPI = QPI1 + QPI2 Next, including the case where QPI is set to 0 in step 3010, the final injection amount QFIN is calculated by adding the final pilot increase QPI to the basic injection amount QBASE obtained in step 2800 (step 3080).

QFIN = QBASE + QPI Then, an injection amount command value VS corresponding to the obtained final injection amount QFIN is obtained and output to the injection amount control actuator drive circuit such as the piezoelectric actuator 100 (step).
3090).

As described above, in the third embodiment, the basic correction amount QPI0 based on the learning value NFG obtained by smoothing and averaging the governor pattern movement integral correction amount NI in the idling stable state.
The pilot injection correction amount QPI2 calculated from the QPI0 and the coefficient f (Ne) determined by the rotational speed assuming the leakage amount ΔQloss from the piezoelectric actuator, and the final injection amount QFIN in the traveling state is corrected and calculated as needed. ing. For example, if the leakage amount ΔQloss increases, QP
I2 is increased, and as a result, the final pilot increase QPI is increased. Therefore, according to the fuel injection control device of the third embodiment, even if the leakage amount ΔQloss from the piezoelectric actuator at the time of execution of the pilot injection fluctuates due to the combustion characteristics, the final pilot increase amount is calculated in consideration of the fluctuation. Therefore, the injection amount most desirable for the diesel engine can be actually controlled to be injected into the combustion chamber irrespective of the change in the fuel property. For this reason, the fuel injection control device of the present embodiment is a pilot injection type fuel injection control device capable of always maintaining stable output performance while avoiding an unexpected situation such as a decrease in torque.

Further, the present embodiment has the same advantage as the first embodiment in that the sensor can be configured at a low cost because no special sensor is required.
In addition, since the final pilot increase QPI is accurately calculated according to each operating condition, the generation of black smoke and deterioration of drivability during excessive EGR in exhaust gas recirculation control, which has recently been adopted as an exhaust gas countermeasure, is avoided. It is possible to

In the embodiment described above, the case where the governor pattern is moved in the rotation axis direction has been described. However, the same applies even when the injection amount (governor pattern) is obtained by a calculation formula and the governor pattern is moved in the injection amount axis direction. effective. That is, the idle governor pattern is set to, for example, Q = aNe + b
(Q: injection amount, a (<0), b (> 0), constant), the idle governor pattern can be changed by increasing or decreasing the constant b proportionally or integrally according to the error from the target value. The fuel amount may be corrected based on the integral correction amount and each operating state by calculating the integral correction amount in accordance with the error by moving the fuel injection member in the injection amount axis direction.

If the basic correction amount is calculated by a predetermined calculation formula or the like based on the error ΔNIDL of the rotation speed obtained in step 560 (step 1560) instead of the basic correction amount calculation process from step 600 to step 730, the correction is performed. The processing content required for calculating the amount becomes easy, and the execution time of the processing can be shortened.

Effects of the Invention As described above in detail including the embodiments, the fuel injection control device of the present invention obtains the reference correction amount of the fuel injection amount when the diesel engine is in a stable idle state, and when the diesel engine is not in the idle state, The reference correction amount is corrected based on the leak correction value that changes according to the rotation speed of the diesel engine and becomes smaller as the rotation speed increases at or above a predetermined rotation speed, and the corrected reference correction amount is used at that time. Is calculated. For this reason, regardless of the aging of the engine, the change in the fuel properties, the variation in the adjustment of the injection pump, etc., a fuel injection control device capable of always injecting the most desirable required injection amount for the diesel engine into the combustion chamber. Become. Therefore, it is possible to reliably avoid a decrease in output or generation of smoke based on a decrease or increase in the amount of fuel injected into the combustion chamber, a decrease in torque, and the like, and always maintain a stable output performance. Further, since a special sensor or the like is not required, the fuel injection control device has a high performance and is inexpensive.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a basic configuration of the present invention, FIG. 2 is a system configuration diagram of an embodiment of the present invention, FIG. 3 is a block diagram of an electronic control unit thereof, FIG. 7 (a), 7 (b), 8, 10 and 12 show the first embodiment.
FIG. 5 is a signal waveform diagram of a rotation speed sensor, FIG. 9 is a graph showing a function of cooling water temperature, and FIG. 11 is a prospective governor for the cooling water temperature. FIG. 13 is a graph showing a relationship between pattern correction amounts, FIG. 13 is a graph showing a relationship between an error between a target rotation speed and an actual rotation speed, and an integration correction amount, and FIG. 14 is a pattern in which a governor pattern is translated in a rotation axis direction. Graph, number
FIG. 15 is a graph showing the relationship between the rotational speed and the correction coefficient, and FIG.
FIG. 17 and FIG. 17 are explanatory diagrams used for obtaining the relationship, and FIG.
(A) and (b) are flowcharts of processing performed in the second embodiment, FIG. 19 is a two-dimensional map diagram in which correction coefficients are determined, and FIG. 20 is a system configuration of the third embodiment. Fig. 21,
FIGS. (A) and (b) are flowcharts of processing executed in the third embodiment, FIG. 22 is a graph showing a relationship between a spill angle and an injection amount, and FIG. 23 is used for obtaining a correction amount. FIG. 1 ... fuel injection pump, 2 ... diesel engine 20 ... electromagnetic spill valve, 25 ... rotational speed sensor 54 ... water temperature sensor, 55a ... air conditioner 55 ... air conditioner switch 56 ... power steering switch 57 ... neutral Switch 58 58 Vehicle speed sensor 59 Electromagnetic clutch for air conditioner 60 Electronic control unit 100 Piezoelectric actuator 103 Piezoelectric element

Continuation of the front page (72) Inventor Ken Ando 1 Toyota Town, Toyota City, Aichi Prefecture Inside Toyota Motor Corporation (56) References JP-A-59-115437 (JP, A) JP-A-63-61744 (JP, A) JP-A-56-75928 (JP, A) JP-A-62-32254 (JP, A) JP-A-57-24428 (JP, A)

Claims (1)

(57) [Claims] 1. An operation state detection means for detecting an operation state including at least a rotation speed of a diesel engine, and an injection amount calculation for calculating a fuel injection amount to be injected into the diesel engine in accordance with a detection result of the operation state detection means. Means, state determination means for determining whether or not the diesel engine is in an idle state, and calculating a deviation of a rotational speed of the diesel engine from a target idle speed when the diesel engine is in an idle state. Deviation calculating means, when the diesel engine is in an idle state, a reference correction amount calculating means for calculating a reference correction amount for increasing or decreasing the fuel injection amount in order to reduce the deviation to a predetermined value or less, and the injection amount calculating means Injection control means for controlling injection of an amount of fuel obtained by correcting the fuel injection amount calculated by the above with the reference correction amount. In the fuel injection control device for a diesel engine, when the diesel engine is not in an idle state, based on a leak correction value that changes according to the rotation speed of the diesel engine and becomes smaller as the rotation speed becomes higher above a predetermined rotation speed. A fuel injection control device for a diesel engine, further comprising a leak correction unit for correcting the reference correction amount.