JP4882883B2 - Fuel injection control device and fuel injection system using the same - Google Patents

Description

  The present invention relates to a fuel injection control device that controls an injection amount of a fuel injection valve that performs multi-stage injection before main injection, and a fuel injection system using the same.

  2. Description of the Related Art Conventionally, an injection control device that controls an injection amount of a fuel injection valve that performs one or more micro injections (also referred to as pilot injection) before main injection is known (see, for example, Patent Document 1). In Patent Document 1, the minute injection amount per correction is corrected based on the average engine rotation speed when n micro injections are performed.

  By the way, in the case of multi-stage injection in which micro injection is performed a plurality of times before main injection, pressure pulsation occurs when the fuel injection valve is closed and the injection is once ended in the first injection of multi-stage injection. When the fuel injection valve performs the second fuel injection of the multi-stage injection, the fuel pressure supplied to the fuel injection valve fluctuates due to the influence of the pressure pulsation generated by the first injection. As a result, even if the fuel injection at each stage of the multi-stage injection injected before the main injection is controlled by, for example, an injection drive signal having a predetermined pulse width, the second injection amount is a target injection corresponding to the predetermined pulse width. May deviate from the amount.

Therefore, it is conceivable to perform single-stage injection and two-stage injection, learn the second injection quantity of multi-stage injection from the difference between the respective injection quantities, and correct the injection quantity according to the learned value.
In Patent Document 1, when two or more pilot injections are performed before the main injection, the influence of the pressure pulsation generated by the first injection on the second injection is not taken into consideration. Learning the second injection amount of the multistage injection from the difference in injection amount from the injection is not considered.
JP 2003-254139 A

  However, there are few opportunities to learn the pressure region on the high pressure side within the pressure range that can be learned according to the engine operating state, so the single-stage injection amount is learned in the entire pressure region of the learning range, and the learning of the single-stage injection amount ends. If an attempt is made to learn the two-stage injection amount that is injected twice after that, the time required to complete the learning of the single-stage injection amount in the entire pressure region of the learning range becomes longer. Thereby, since the start of learning of the two-stage injection amount is delayed, it takes a long time to complete learning of both the single-stage injection amount and the two-stage injection amount in each pressure region. As a result, in each pressure region, for example, it takes a long time to correct the pulse width of the second injection drive signal based on the learning value.

  Further, when learning the two-stage injection amount after learning of the single-stage injection amount is completed in the entire pressure region of the learning range, the learning time interval between the single-stage injection and the two-stage injection becomes long. As a result, if the single-stage injection amount has changed due to changes over time when learning the second-stage injection amount, the accuracy of learning the second injection amount from the difference between the two-stage injection amount and the single-stage injection amount is improved. descend.

  The present invention has been made to solve the above-described problem, and shortens the time required for completing learning of the single-stage injection amount and the two-stage injection amount in the pressure region as much as possible, and improves the learning accuracy. It is an object of the present invention to provide an injection control device and a fuel injection system using the same.

In the invention according to claim 1, in a learning range constituted by a plurality of learning regions in which each learning region has at least one pressure region, there is a pressure region in which the single stage injection amount is not learned in the predetermined learning region. In this case, the learning of the unlearned single-stage injection amount is preferentially selected, and when the learning of the single-stage injection amount ends in the predetermined learning region and there is a pressure region in which the two-stage injection amount is not learned, the learning is Select learning of the two-stage injection amount.

  That is, the learning of the single-stage injection amount is first selected in the predetermined learning region, and then the learning of the single-stage injection amount in the other learning region is not shifted to the second learning amount in the same learning region. Select learning.

  As described above, learning of both the single-stage injection quantity and the two-stage injection quantity is completed in the predetermined learning area, and learning of the single-stage injection quantity and the two-stage injection quantity is selected with the other learning area as the predetermined learning area. As a result, in the pressure region of the learning region, the time required to complete learning of both the single-stage injection amount and the two-stage injection amount is shortened.

  Compared to learning the single-stage injection amount after learning the single-stage injection amount in the entire pressure region of the learning range, the single-stage injection amount is learned after learning the single-stage injection amount in the same pressure region of the learning region. The time interval until learning is shortened. As a result, the two-stage injection quantity in the same pressure region can be learned before the single-stage injection quantity changes due to changes over time, etc., so the second injection quantity is highly accurate from the difference between the two-stage injection quantity and the single-stage injection quantity. To learn.

In the invention according to claim 2, in the learning range constituted by a plurality of learning regions in which each learning region has at least one pressure region, learning of the single-stage injection amount ends in the predetermined learning region, and the two-stage injection amount If there is an unlearned pressure region, the learning of the unlearned two-stage injection amount is preferentially selected, and if there is an unlearned pressure region in the predetermined learning region, the unlearned Select learning of single-stage injection amount.

  Thus, when learning of the single-stage injection amount in the pressure region ends, learning of the single-stage injection amount in the other pressure region is not selected, but learning of the unlearned two-stage injection amount in the same pressure region is prioritized. Therefore, in each pressure region, the time required for learning of both the single-stage injection amount and the two-stage injection amount is shortened.

Compared to learning the single-stage injection amount after learning the single-stage injection amount in the entire pressure region of the learning range, the multi-stage injection amount is learned after learning the single-stage injection amount in the same pressure region of the learning region. The time interval until is shortened. As a result, the two-stage injection quantity in the same pressure region can be learned before the single-stage injection quantity changes due to changes over time, etc., so the second injection quantity is highly accurate from the difference between the two-stage injection quantity and the single-stage injection quantity. To learn.
In the first and second aspects of the invention, the injection amount of the single-stage injection amount or the two-stage injection amount selected in the pressure region is learned, and the injection amount between the learned two-stage injection amount and single-stage injection amount The injection amount characteristic of the fuel injection valve is corrected based on the difference. Thereby, for example, the injection amount characteristic of the fuel injection valve can be corrected from the pressure region where learning of the single-stage injection amount and the two-stage injection amount has been completed a predetermined number of times. Therefore, the injection amount of the fuel injection valve can be controlled with high accuracy based on the corrected injection amount characteristic.

  By the way, in a normal engine operation state, learning on the high pressure side has fewer learning opportunities than learning on the low pressure side, so there is an opportunity to judge learning on the high pressure side, and it takes a long time to complete the learning of the entire learning range. is required.

  Therefore, as in the inventions of claims 3 and 5, by determining learning with priority from the high pressure side, the learning of the single-stage injection amount and the two-stage injection amount is completed in the entire learning range. The time required is shortened.

  In the inventions according to claims 4 and 5, the learning is preferentially determined in the learning region (also referred to as the priority learning region) in which multistage injection is frequently used. As a result, in the priority learning region, the time required to complete learning of both the single-stage injection amount and the two-stage injection amount is shortened.

  The functions of the plurality of means provided in the present invention are realized by hardware resources whose functions are specified by the configuration itself, hardware resources whose functions are specified by a program, or a combination thereof. The functions of the plurality of means are not limited to those realized by hardware resources that are physically independent of each other.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
A fuel injection system according to an embodiment of the present invention is shown in FIG.
(Fuel injection system 10)
The accumulator fuel injection system 10 includes a feed pump 14, a high-pressure pump 16, a common rail 20, a pressure sensor 22, a pressure reducing valve 24, a fuel injection valve 30, an electronic control unit (ECU) 40, an electronic drive unit (Electronic). Driving unit (EDU) 42 and the like, and fuel is injected from the fuel injection valve 30 into each cylinder of the four-cylinder diesel engine 50. To avoid complexity of the figure, it shows only the driving signal line to the one fuel injection valve 30 from EDU42 in FIG.

  The feed pump 14 sucks fuel from the fuel tank 12 and supplies it to a high-pressure pump 16 that is a fuel supply pump. The high-pressure pump 16 is a known pump that pressurizes the fuel sucked into the pressurizing chamber when the plunger reciprocates as the cam of the camshaft rotates. By controlling the current value supplied to the metering valve 18 of the high-pressure pump 16 by the ECU 40, the fuel suction amount that the high-pressure pump 16 sucks in the suction stroke is metered. Then, by adjusting the fuel intake amount, the fuel discharge amount of the high-pressure pump 16 is adjusted.

  The common rail 20 accumulates fuel pumped by the high-pressure pump 16 and holds the fuel pressure at a predetermined high pressure according to the engine operating state. The pressure sensor 22 as pressure detecting means detects the fuel pressure inside the common rail 20 and outputs it to the ECU 40.

  The pressure reducing valve 24 opens to discharge the fuel inside the common rail 20 to the return passage 100 on the low pressure side. The pressure reducing valve 24 is, for example, a known art in which a load of a spring is applied to a valve member in a valve closing direction, and an electromagnetic drive unit such as a coil is energized to lift and open the valve member against the spring load. It is a solenoid valve. The valve opening time of the pressure reducing valve 24 becomes longer according to the pulse width (energizing time) of the energization pulse energized to the pressure reducing valve 24.

  The fuel injection valve 30 is installed in each cylinder of the four-cylinder diesel engine 50, and injects the fuel accumulated in the common rail 20 into the cylinder. The fuel injection valve 30 performs multi-stage injection including pilot injection and main injection in one combustion stroke of the diesel engine 50. The fuel injection valve 30 is a known electromagnetically driven valve that controls the fuel injection amount by controlling the pressure in a control chamber that applies fuel pressure to the nozzle needle in the valve closing direction.

  The ECU 40 as a fuel injection control device is composed of a microcomputer (microcomputer) centered on a rewritable nonvolatile memory such as a CPU, ROM, RAM, flash memory or the like. A control program that causes the ECU 40 to function as a learning determination unit, a learning selection unit, a learning unit, and a correction unit is stored in a storage device such as a ROM or a flash memory.

  The ECU 40 includes a rotational speed sensor that detects an engine rotational speed (NE), an accelerator sensor that detects an opening degree (ACC) of an accelerator pedal, a crank angle sensor that detects a crank angle (CA), a temperature sensor, a pressure sensor 22, and the like. The operating state of the diesel engine 50 is acquired from detection signals of various sensors. The ECU 40 controls energization to the metering valve 18, the pressure reducing valve 24, the fuel injection valve 30 and the like based on the acquired engine operating state in order to control the diesel engine 50 to an optimal operating state.

  Further, the ECU 40 stores the discharge amount characteristic of the discharge amount of the high-pressure pump 16 with respect to the energization amount to the metering valve 18 as a map in the flash memory according to the engine operating state. The ECU 40 feedback-controls energization to the metering valve 18 so that the common rail pressure acquired from the pressure sensor 22 becomes the target common rail pressure based on the discharge amount characteristic of the high-pressure pump 16 stored in the flash memory.

  Further, the ECU 40 stores, as an injection characteristic map, a flash memory that stores the relationship between the energization amount to the electromagnetic drive unit of the fuel injection valve 30 and the injection amount of the fuel injection valve 30 according to the engine operating state such as the common rail pressure. . The injection amount characteristic map of the present embodiment represents the relationship between the pulse width of the injection drive signal and the injection amount. The ECU 40 controls the injection of each stage of the multi-stage injection including the main injection of the fuel injection valve 30 based on the injection amount characteristic of the fuel injection valve 30 stored in the flash memory.

The EDU 42 is a drive device for supplying drive current or drive voltage to the pressure reducing valve 24 and the fuel injection valve 30 based on a control signal output from the ECU 40.
(Injection amount learning)
Next, the injection amount learning of the multistage injection performed before the main injection will be described. The injection amount learning routine of FIG. 3 is executed, for example, when the engine is operating with the engine speed decreasing.

  In S300, the ECU 40 determines whether the learning condition is satisfied. For example, the ECU 40 determines whether the accelerator is off and the fuel injection valve 30 is in the non-injection state, the engine speed is decreasing at a constant rate, and the diesel engine 50 is operating for a predetermined time or more. If the learning condition is not satisfied, the ECU 40 ends this routine.

  When the learning condition is satisfied in S300, the ECU 40 can calculate the upper limit value of the common rail pressure that can be learned by driving the high-pressure pump 16 from the current value of the engine speed and the common rail pressure in S302, and can learn the injection amount. A pressure range (hereinafter also simply referred to as “learning range”) is calculated. The calculated upper limit value of the common rail pressure is equal to or lower than the maximum pressure required as the common rail pressure. The lower limit value of the learning range is the minimum pressure required for operating the diesel engine 50, and is a constant value regardless of the engine operating state in the present embodiment.

  In S304, the ECU 40 selects the pressure region corresponding to the learning common rail to be learned in the learning range, and whether to learn the single-stage injection amount (SQL) or the two-stage injection amount (WSQL) at the learning common rail pressure. This routine is terminated. The selection of the pressure region to be learned in the learning range and whether to learn the single-stage injection quantity or the two-stage injection quantity is processed by a learning selection routine shown in FIG. 4 or FIG. Details will be described later.

(Learning selection pattern 1)
The outline of the injection amount learning selection pattern 1 executed in S304 of the injection amount learning routine of FIG. 3 will be described with reference to FIG. The numbers shown in FIG. 1 indicate the order in which learning is determined. In FIG. 1, the learning region described in the claims is composed of one pressure region.

  The learning range calculated in S302 of the injection amount learning routine of FIG. 3 includes a plurality of pressure regions (Pn, Pn + 1, Pn + 2, Pn + 3...) From the high pressure side. Pn is the upper limit value of the common rail pressure calculated in S302 of FIG. When learning each pressure region in the learning range shown in FIG. 1, the ECU 40 controls the current value supplied to the metering valve 18 of the high-pressure pump 16 and learns by metering the discharge amount of the high-pressure pump 16. The pressure is adjusted to the pressure region.

  Then, the ECU 40 proceeds from the high pressure side to the low pressure side, and if the learning of the single-stage injection amount has been completed but there is a pressure region in which the two-stage injection amount has not been learned, the process does not shift to another pressure region. The two-stage injection amount learning in the same pressure region where the learning of the single-stage injection amount has been completed is selected.

If learning of the single-stage injection amount has ended but there is no pressure region in which the two-stage injection amount has not been learned, the ECU 40 selects learning of the single-stage injection amount in the pressure region in which the single-stage injection amount has not been learned. To do.
Thus, in the learning selection pattern 1, if the two-stage injection amount is unlearned in the same pressure region where the learning of the single-stage injection amount has ended, the learning determination is not transferred to another pressure region and the two steps in the same pressure region are performed. Select with priority on learning of injection quantity. As a result, in the pressure range of the learning range, the time required to complete learning of both the single-stage injection amount and the two-stage injection amount is shortened.

  In each pressure region, the time interval from the end of learning of the single-stage injection amount to the end of learning of the two-stage injection amount is shortened. As a result, the two-stage injection quantity in the same pressure region can be learned before the single-stage injection quantity changes due to changes over time, etc., so the second injection quantity is highly accurate from the difference between the two-stage injection quantity and the single-stage injection quantity. To learn. As a result, even if the second injection amount in the second-stage injection deviates from the target value due to the pressure pulsation generated as a result of the first injection, the ECU 40 corrects the injection amount characteristic map based on the learned value and corrects the injection drive. By controlling the fuel injection valve 30 with a pulse, the two-stage injection amount before the main injection can be brought close to the target value.

  Further, learning of the single-stage injection amount and the two-stage injection amount is selected from the pressure region on the high-pressure side where there are few learning opportunities, and the learning of the selected pressure region is completed by driving the high-pressure pump 16, so the entire learning range The time required to complete learning of the single-stage injection amount and the two-stage injection amount is shortened.

(Learning selection routine 1)
Next, specific processing by the learning selection pattern 1 shown in FIG. 1 will be described based on the learning selection routine 1 of FIG. The learning selection routine 1 in FIG. 4 is executed in S304 in FIG. The SQL execution flag, WSQL execution flag, and learnable common rail pressure absence flag described below are set to 0 as initial values.

First, in S320, the ECU 40 sets the maximum pressure (Pn) to the learning common rail pressure within the learnable pressure range shown in FIG.
In S322, the ECU 40 determines whether there is a cylinder that has not yet learned the single-stage injection amount in the predetermined pressure region represented by the learning common rail pressure. If there is a cylinder that has not yet learned the single-stage injection amount, the ECU 40 sets the SQL execution flag to 1 in S324 and ends this routine. If the SQL execution flag is 1, it indicates that learning of the single-stage injection amount in the corresponding pressure region is selected, and if it is 0, it indicates that learning of the single-stage injection amount is not selected.

  If there is no cylinder that has not yet learned the single-stage injection amount in the predetermined pressure region, the ECU 40 determines in S326 whether there is a cylinder that has not yet learned the second-stage injection amount. If there is a cylinder that has not learned the two-stage injection amount, the ECU 40 sets the WSQL execution flag to 1 in S328 and ends this routine. If the WSQL execution flag is 1, it indicates that learning of the two-stage injection amount in the corresponding pressure region is selected, and if it is 0, it indicates that learning of the two-stage injection amount is not selected.

  If there is no cylinder that has not yet learned the single-stage injection amount or the two-stage injection amount in the predetermined pressure region, in S330, the ECU 40 determines whether a pressure region lower than the current predetermined pressure region exists within the learned pressure range. To do. If it does not exist, learning of the single-stage injection quantity and the two-stage injection quantity has been completed in the entire pressure region within the learning pressure range of FIG. In this case, in S332, the ECU 40 sets the learnable common rail pressure no flag to 1 and ends this routine. The learnable common rail pressure no flag indicates that there is no learnable common rail pressure if it is 1, and that there is a learnable common rail pressure if it is 0.

  If there is a pressure region lower than the current predetermined pressure region in the learning range, in S334, the ECU 40 newly sets the common rail pressure in the high pressure region next to the current predetermined pressure region to the learning common rail pressure. , The process proceeds to S322.

(Learning, injection amount correction)
FIG. 5 shows the injection amount learning of the single-stage injection amount or the two-stage injection amount selected in the pressure region corresponding to the learning common rail pressure selected in the learning selection routine 1 of FIG. 4, and the injection is performed based on the learning value. This is a routine for correcting the amount.

  In S340, the ECU 40 drives the fuel injection valve 30 in the pressure region selected in the learning selection routine 1, and performs injection amount learning of an unlearned single-stage injection amount or two-stage injection amount. Specifically, the ECU 40 calculates a variation amount of the engine torque from the variation number of the engine speed that changes in synchronization with the injection drive signal, and calculates an injection amount from the variation amount of the engine torque.

  In S342, the ECU 40 determines whether learning of the single-stage injection amount and the second-stage injection amount is completed in the same pressure region, and the number of learning exceeds a predetermined number K. If the learning number does not exceed the predetermined number K, the ECU 40 ends this routine.

  When the learning number exceeds the predetermined number K, in S344, the ECU 40 calculates the average of the learning values of the single-stage injection amount and the two-stage injection amount. The second injection amount can be calculated from the difference between the averaged two-stage injection amount and the single-stage injection amount.

  In S346, when the learned second injection amount is different from the target injection amount, the ECU 40 corrects the injection amount characteristic map stored in the flash memory on the basis of the injection amount deviation amount, and sets the actual injection amount. Approach the target injection amount.

(Learning selection pattern 2)
Based on FIG. 6, the outline of the learning selection pattern 2 having an injection amount different from that of the learning selection pattern 1 will be described. The numbers shown in FIG. 6 indicate the order in which learning is determined. In FIG. 6, the learning area (priority learning area) for which priority is given to the learning described in the claims is composed of a plurality of pressure areas. The learning area described in (1) is composed of all pressure areas other than the priority learning area .

First, the ECU 40 calculates a learning range in the current engine operating state. The learning range includes a priority learning area (Pn, Pn + 1, Pn + 2) and all pressure areas (Pm, Pm + 1,...) Other than the priority learning area. The priority learning region is a pressure range in which two-stage injection is frequently used, and represents a pressure range in a normal travel region between 0 km / h and 60 km / h, for example.

  Then, if there is a pressure region in which the single-stage injection amount is not learned from the high pressure side to the low pressure side in the priority learning region, the ECU 40 selects learning of the single-stage injection amount in the pressure region. When the learning of the single-stage injection amount ends in the priority learning region, the ECU 40 does not shift to the learning determination of the single-stage injection amount other than the priority learning region, and proceeds from the high pressure side to the low pressure side of the priority learning region. If there is a pressure region in which the step injection amount is not learned, learning of the two-step injection amount in the priority learning region is selected.

When the learning of the single-stage injection amount and the two-stage injection amount is completed in the entire pressure region of the priority learning region, the ECU 40 decreases the pressure from the high pressure side in the learning region configured by all the pressure regions other than the priority learning region in the learning range. If there is a pressure region in which the single-stage injection amount is not learned toward the side, learning of the single-stage injection amount in the pressure region is selected.

  When the learning of the single-stage injection amount is completed in the pressure region other than the priority learning region, the ECU 40 proceeds to 2 in the pressure region if there is a pressure region in which the two-stage injection amount is not learned from the high pressure side to the low pressure side. Select learning of stage injection quantity.

  Thus, since learning of both the single-stage injection amount and the two-stage injection amount is completed in the priority learning region and learning of the single-stage injection amount and the two-stage injection amount is selected in another pressure region, the priority learning region , The time required for learning of both the single-stage injection amount and the two-stage injection amount is shortened.

  In addition, since learning in the priority learning region is prioritized, the time interval from learning the single-stage injection amount to learning the two-stage injection amount in the same pressure region in the priority learning region is shortened. As a result, the two-stage injection quantity in the same pressure region can be learned before the single-stage injection quantity changes due to changes over time, etc., so the second injection quantity is highly accurate from the difference between the two-stage injection quantity and the single-stage injection quantity. To learn.

  As a result, even if the second injection amount in the second stage injection deviates from the target value due to the pressure pulsation resulting from the first injection, the injection amount characteristic map is corrected based on the learning value, and the fuel is generated by the corrected injection drive pulse. By controlling the injection valve 30, the two-stage injection amount before the main injection can be brought close to the target value.

  When learning in the priority learning region is completed, learning of the single-stage injection amount and the two-stage injection amount is selected from the pressure region on the high pressure side where the learning opportunity is small in the learning range other than the priority learning region, and the high pressure pump 16 is driven. Since the common rail pressure is set in the selected pressure region and the learning of the single-stage injection amount and the two-stage injection amount is completed, it is necessary to complete the learning of the single-stage injection amount and the two-stage injection amount in the entire learning range. Time is shortened.

(Learning selection routine 2)
The learning selection routine 2 in FIG. 7 is executed in S304 in FIG. 3 instead of the learning selection routine 1 in FIG. The priority learning region end flag, the SQL execution flag, the WSQL execution flag, and the learnable common rail pressure no flag described below are set to 0 as initial values.

  In S350, the ECU 40 determines whether the priority learning area end flag is 1. If the priority learning region end flag is not 1, the ECU 40 sets the priority learning region of the learnable pressure range as the learning execution region in S352. If the priority learning region end flag is 1, in S354, the ECU 40 sets all pressure regions other than the priority learning region of the learnable pressure range as the learning execution region.

  If the priority learning region end flag is 1, it means that all learning of the single-stage injection amount and the two-stage injection amount has been completed in the priority learning region, and if it is 0, the single-stage injection amount or two-stage injection in the priority learning region. This indicates that there is a pressure region in which the injection amount is not learned.

  Next, in S356, the ECU 40 sets the step number variable n to 1, and in S358 sets the maximum pressure to the learning common rail pressure in the learning execution region. If the stage number variable n is 1, it represents a single stage, and if it is 2, it represents two stages.

  In S360, the ECU 40 determines whether there is a cylinder that has not yet learned the single-stage or two-stage injection amount indicated by the stage number variable n in the predetermined pressure region represented by the learning common rail. If there is a cylinder that has not yet learned the single-stage injection quantity or the two-stage injection quantity in the predetermined pressure region, the ECU 40 sets the SQL execution flag to 1 when the stage number variable n is n = 1 in S362, and n = 2 At this time, the WSQL execution flag is set to 1 and this routine is terminated. If the SQL execution flag and the WSQL execution flag are 1, it indicates that learning of the number of stages indicated by the flag is selected in the corresponding pressure region, and if 0, learning of the number of stages indicated by the flag is not selected. To express.

  If there is no cylinder that has not yet learned the single-stage injection amount or the two-stage injection amount in the predetermined pressure region, in S364, the ECU 40 determines whether a pressure region lower than the current learning common rail pressure exists in the current learning execution region. Determine. If present, in S366, the ECU 40 sets the common rail pressure corresponding to the next higher pressure region after the current learning common rail pressure in the current learning execution region as the learning common rail pressure, and the process proceeds to S360.

  If there is no pressure region lower than the current learning common rail pressure in the current learning execution region, the ECU 40 determines in step S368 whether the stage number variable n is n = 1. If n = 1, it indicates that learning of the single-stage injection amount is completed in the learning execution region, and that the second-stage injection amount is not learned. Therefore, the ECU 40 sets the step number variable n to 2 in S370, and shifts the processing to S358 in order to learn the two-stage injection amount from the maximum pressure in the learning execution region.

If n is not 1 in S368, it indicates that learning of the single-stage injection quantity and the two-stage injection quantity has been completed in the learning execution region.
In this case, in S372, the ECU 40 determines whether the current learning execution area is a priority learning area. If the current learning execution area is not the priority learning area, the ECU 40 sets the learnable common rail pressure no flag to 1 in S374 and ends this routine. This indicates that learning of the single-stage injection amount and the two-stage injection amount has been completed in the entire pressure region of the learnable pressure range.

If the current learning execution area is the priority learning area, the ECU 40 sets the priority learning area end flag to 1 in S376, and the process proceeds to S350.
When the learning selection routine 2 in FIG. 7 ends, the ECU 40 executes a learning execution and injection amount correction routine shown in FIG.

(Learning selection pattern 3)
Based on FIG. 8, the outline of the learning selection pattern 3 having an injection amount different from the learning selection patterns 1 and 2 will be described. The numbers shown in FIG. 8 indicate the order in which learning is determined. In FIG. 8, the priority learning region is composed of a plurality of pressure regions, and in the learning range other than the priority learning region, the learning region described in the claims is composed of one pressure region.

  First, the ECU 40 calculates a learning range in the current engine operating state. The learning range includes a priority learning region (Pn, Pn + 1, Pn + 2) in which two-stage injection is frequently used, and all pressure regions (Pm, Pm + 1,...) Other than the priority learning region.

  Then, in the priority learning region, the ECU 40 proceeds to another pressure region from the high pressure side toward the low pressure side if there is a pressure region in which the learning of the single-stage injection amount has ended but the two-stage injection amount has not yet been learned. The two-stage injection amount learning in the same pressure region where the learning of the single-stage injection amount has ended is selected without shifting to the process.

If learning of the single-stage injection amount has ended in the priority learning region but there is no pressure region in which the two-stage injection amount has not yet been learned, the ECU 40 determines that learning of the priority learning region has ended.
When the learning of the priority learning region is completed, the ECU 40 has completed learning of the single-stage injection amount from the high pressure side toward the low pressure side in the pressure region other than the priority learning region within the learning range, but the two-stage injection amount. If there is an unlearned pressure region, the process is not transferred to another pressure region, and the two-stage injection amount learning in the same pressure region where the learning of the single-stage injection amount has been completed is selected.

The ECU 40 determines that learning of the learnable pressure range has ended if learning of the single-stage injection amount has ended but there is no pressure region in which the two-stage injection amount has not yet been learned.
[Other Embodiments]
In the above embodiment, one priority learning area is set. On the other hand, a plurality of priority learning areas may be set, and a priority order may be set among them. Further, the priority learning area may be composed of one pressure area.

  In the above embodiment, an example in which two-stage injection is performed before main injection has been described. In contrast, three or more stages of injection may be performed before the main injection. Also in this case, it is possible to estimate the influence of the pressure pulsation generated by the first injection on the second and subsequent injection amounts by learning the single-stage injection amount and the second-stage injection amount.

  As described above, the present invention is not limited to the above-described embodiment, and can be applied to various embodiments without departing from the gist thereof.

Explanatory drawing which shows the learning selection pattern 1 by this embodiment. The block diagram which shows a fuel-injection system. The flowchart which shows the injection quantity learning routine. The flowchart which shows the learning selection routine 1. The flowchart which shows learning implementation and the injection quantity correction | amendment routine. Explanatory drawing which shows the learning selection pattern 2. FIG. 7 is a flowchart showing a learning selection routine 2; Explanatory drawing which shows the learning selection pattern 3. FIG.

Explanation of symbols

10: fuel injection system, 16: high pressure pump (fuel supply pump), 20: common rail, 22: pressure sensor (pressure detection means), 30: fuel injection valve, 40: ECU (fuel injection control device, learning determination means, learning) (Selection means, learning means, correction means)

Claims (6)

In a fuel injection control device that controls the injection amount of a fuel injection valve that performs multi-stage injection before main injection,
It is determined whether there is a pressure region in which a single stage injection amount has not yet been learned in a predetermined learning region of a learning range configured by a plurality of learning regions each having at least one pressure region, and the predetermined learning Learning determination means for determining whether or not the pressure region that has not learned the second-stage injection amount exists when learning of the single-stage injection amount is finished in the region;
Based on the determination result of the learning determination means, when there is the unlearned pressure region in the predetermined learning region, priority is given to learning of the unlearned single-stage injection amount, Learning selecting means for ending learning of the single-stage injection amount in a predetermined learning region and selecting learning of the two-stage injection amount that has not been learned when the pressure region that has not learned the second-stage injection amount exists; ,
Learning means for learning the injection quantity of the single-stage injection quantity or the two-stage injection quantity selected in the pressure region;
When learning of the single-stage injection quantity and the two-stage injection quantity ends in the pressure region, the second injection quantity is calculated from the difference in the learning injection quantity between the two-stage injection quantity and the single-stage injection quantity, and is calculated Corrective means for correcting the injection amount characteristic of the fuel injection valve based on the deviation amount of the injection amount when the second injection amount is different from the target injection amount;
A fuel injection control device comprising: In a fuel injection control device that controls the injection amount of a fuel injection valve that performs multi-stage injection before main injection,
The learning of the single-stage injection amount ends in the predetermined learning region of the learning range constituted by a plurality of learning regions each having at least one pressure region, and the pressure region where the two-stage injection amount is not learned exists. Learning determination means for determining whether or not there is the pressure region in which the single-stage injection amount has not been learned in the predetermined learning region,
Based on the determination result of the learning determination means, when the learning of the single-stage injection amount is finished in the predetermined learning region and the pressure region where the second-stage injection amount is not learned exists, the second-stage injection amount is not set. The learning of the two-stage injection amount in the pressure region of learning is selected with priority, the learning of the single-stage injection amount ends in the predetermined learning region, and the pressure region where the two-stage injection amount is not learned exists Without the learning selection means for selecting learning of the unlearned single-stage injection amount when the pressure region that has not yet learned the single-stage injection amount exists in the predetermined learning region,
Learning means for learning the injection quantity of the single-stage injection quantity or the two-stage injection quantity selected in the pressure region;
When learning of the single-stage injection quantity and the two-stage injection quantity ends in the pressure region, the second injection quantity is calculated from the difference in the learning injection quantity between the two-stage injection quantity and the single-stage injection quantity, and is calculated Corrective means for correcting the injection amount characteristic of the fuel injection valve based on the deviation amount of the injection amount when the second injection amount is different from the target injection amount;
A fuel injection control device comprising:   3. The fuel injection control device according to claim 1, wherein the learning determination unit makes a determination from the high pressure side toward the low pressure side in the learning range.   3. The fuel injection control device according to claim 1, wherein the learning determination unit preferentially determines learning in the learning region that frequently uses the multistage injection. 4.   The learning determination unit is configured to determine from the high pressure side to the low pressure side in the learning region in which learning is prioritized and in the learning range other than the learning region in which learning is prioritized. Item 5. The fuel injection control device according to Item 4. A fuel supply pump that pressurizes and pumps fuel; and
A common rail for accumulating fuel pumped by the fuel supply pump;
A fuel injection valve for injecting fuel accumulated in the common rail; and
A fuel injection control device according to any one of claims 1 to 5,
A fuel injection system comprising:

Images