JP6156203B2 - Pump characteristic learning device for fuel pump - Google Patents

Description

  The present invention relates to a pump characteristic learning apparatus suitable for learning the pump characteristic of a fuel pump that supplies high-pressure fuel to a common rail.

  Conventionally, as a system for injecting fuel into a vehicle engine, high pressure fuel is stored in the common rail by supplying high pressure fuel from the fuel pump to the common rail, and the accumulated high pressure fuel is stored in each engine cylinder. A common rail system for supplying to a fuel injection valve provided is known.

  In the common rail system, the fuel discharge amount from the fuel pump to the common rail is controlled via the discharge amount control valve provided in the fuel pump. A control system for the fuel pump is designed based on the pump characteristics representing the relationship with the quantity.

  Even with this design, the pump characteristics of the fuel pump change over time. Therefore, when the common rail system is actually used, the deviation of the pump characteristics from the design time is detected and the parameters of the control system ( For example, it has been proposed to perform learning control for correcting the control amount of the discharge amount control valve (see, for example, Patent Documents 1 and 2).

  That is, in the learning control described in Patent Document 1, the fuel discharge amount from the fuel pump is gradually increased via the discharge amount control valve, and the common rail pressure (the fuel in the common rail is determined via the pressure reducing valve provided on the common rail). Pressure) to the target pressure. Then, the control amount of the discharge amount control valve at the saturation point at which the control amount of the pressure reducing valve does not change is obtained as the maximum discharge control amount.

  Further, the control amount of the discharge amount control valve when the fuel pump starts to suck the fuel is obtained as the suction start control amount. Then, the pump characteristics are corrected based on these two control amounts (that is, the minimum and maximum control amounts), and the control amount of the discharge amount control valve is set based on the corrected pump characteristics.

  On the other hand, in the learning control described in Patent Document 2, when the common rail pressure is the predetermined learning target pressure, the fuel discharge amount from the fuel pump is increased by changing the control amount of the discharge amount control valve. Let

  At the same time, the common rail pressure is controlled to the learning target pressure via the pressure reducing valve provided on the common rail, and the increase amount of the fuel discharge amount from the fuel pump is calculated from the amount of fuel passing through the pressure reducing valve.

  Based on the fuel increase amount and the control amount of the fuel control valve, the relationship between the fuel discharge amount from the fuel pump and the control amount of the fuel control valve (that is, pump characteristics) is learned.

JP2013-160110A JP 2010-223130 A

  In each of the proposed learning devices, the two control amounts that can control the fuel discharge amount from the fuel pump by the discharge amount control valve, and the change amount of the fuel discharge amount that occurs between the two control amounts. Therefore, the pump characteristics of the fuel pump are learned, and it is assumed that the pump characteristics of the fuel pump can be linearly approximated.

  However, in a fuel pump, the discharge amount control valve has a control amount determined by its energization current and the valve opening (and thus the fuel discharge amount from the fuel pump) does not change in a one-to-one relationship. It varies nonlinearly depending on the shape of the seating surface of the valve body. Further, the shape of the non-linear change characteristic also changes due to the deterioration of each part.

  Therefore, in the proposed learning device, the control amount of the discharge amount control valve and the fuel discharge from the fuel pump are near the maximum control amount or the minimum control amount of the discharge amount control valve, or the control amount of one point that becomes the learning point. Even if the control amount can be corrected so as to correspond to the amount, it is difficult to appropriately correct the control amount at another intermediate point.

  For this reason, the proposed learning device cannot accurately grasp the fuel discharge amount from the fuel pump with respect to the control amount at the intermediate point deviated from the learning point, and the fuel discharge amount from the fuel pump (by extension) There is a problem that the amount of fuel injected into the engine cannot be controlled with high accuracy.

  The present invention has been made in view of these problems, and is a pump characteristic learning device capable of learning the pump characteristics of a fuel pump that supplies high-pressure fuel to a common rail over the entire opening range of the discharge amount control valve. The purpose is to provide.

  The pump characteristic learning device of the present invention is a common rail system including a common rail, a fuel pump, a discharge amount control valve, and a pressure reducing valve, and the control amount of the discharge amount control valve when the engine is operated in a constant speed state. The pump characteristics representing the relationship with the fuel discharge amount from the fuel pump are learned.

The pump characteristic learning device of the present invention is based on the premise that the drive characteristic of the pressure reducing valve is a linear drive characteristic in which the relationship between the fuel discharge amount and the drive amount is linear.
The pressure reducing valve is opened by energization to discharge fuel from the common rail, and the amount of fuel discharged from the pressure reducing valve is normally proportional to the valve opening time that is the driving amount of the pressure reducing valve. Therefore, in general, the common rail pressure reducing valve has a linear drive characteristic.

  Next, the pump characteristic learning device of the present invention includes a decompression start point search means, a discharge amount increase control means, a control amount storage means, a saturation point search means, and a pump characteristic specifying means. Works when the engine is operating at a constant speed.

  Here, the pressure reducing start point searching means gradually opens the pressure reducing valve from the closed state while controlling the fuel pressure in the common rail to the target pressure via the discharge amount control valve, and the discharge amount control valve. Search for a change point where the controlled variable changes. Then, the drive amount of the pressure reducing valve at the change point is stored as the pressure reduction start drive amount.

  Further, the discharge amount increase control means sequentially increases the fuel discharge amount from the fuel pump to the common rail via the discharge amount control valve, and discharges the fuel in the common rail via the pressure reducing valve. Control the fuel pressure to the target pressure.

  The control amount storage means calculates the control amount of the discharge amount control valve and the drive amount of the pressure reducing valve required to control the fuel pressure to the target pressure each time the discharge amount increase control means increases the fuel discharge amount. Remember.

  Further, the saturation point search means enters a saturation region where the discharge amount control valve cannot increase the fuel discharge amount when the drive amount of the pressure reducing valve does not change even when the discharge amount increase control means increases the fuel discharge amount. The operation of the discharge amount increase control means is terminated.

  Thereafter, the saturation point search means gradually changes the control amount of the discharge amount control valve by a change amount smaller than the change amount at the time of increase by the discharge amount increase control means, thereby changing the drive amount of the pressure reducing valve. A saturation point that is a boundary between the discharge amount controllable region and the saturation region is searched.

Further, when the saturation point is searched for, the saturation point searching means stores the drive amount of the pressure reducing valve and the control amount of the discharge amount control valve at the saturation point as the total discharge amount driving amount and the final control amount, respectively.
Then, the pump characteristic specifying unit stores the decompression start drive amount stored in the decompression start point search unit, the total discharge amount drive amount and final control amount stored in the saturation point search unit, and the control amount storage unit. The pump characteristics are specified based on the control amount and the drive amount stored for each increase in the fuel discharge amount.
Specifically, the pump characteristic specifying means is based on a linear characteristic representing the relationship between the drive amount of the pressure reducing valve and the fuel discharge amount from the fuel pump, which is obtained from the pressure reduction start drive amount and the total discharge amount drive amount. The pump characteristics representing the relationship between the control amount of the discharge control valve and the fuel discharge amount from the fuel pump in the entire fuel discharge region

  That is, the fuel discharge amount from the fuel pump is “zero” when the drive amount of the pressure reducing valve is the drive amount for starting the pressure reduction, and “maximum” when the drive amount of the pressure reducing valve is the drive amount at the time of full discharge amount. . Since the pressure reducing valve has a linear drive characteristic, the fuel discharge amount from the fuel pump is obtained by proportional calculation with respect to the change in the drive amount from “zero” to “maximum” of the drive amount of the pressure reducing valve. be able to.

  On the other hand, the control amount of the discharge amount control valve at the saturation point at which the fuel discharge amount from the fuel pump becomes “maximum” is stored as the final control amount. Further, in the region where the fuel discharge amount from the fuel pump is from “zero” to “maximum”, the discharge amount increase control means increases the fuel discharge amount from the fuel pump every time the fuel discharge amount is increased. Thus, the control amount of the discharge amount control valve is stored.

  Therefore, from these stored results, the pump characteristics from when the discharge amount control valve is opened until the final control amount is reached (that is, the relationship between the control amount of the discharge amount control valve and the fuel discharge amount from the fuel pump). ) Can be specified by calculating the fuel discharge amount corresponding to the control amount of the discharge amount control valve by proportional calculation based on the drive amount of the pressure reducing valve.

Therefore, according to the pump characteristic learning device of the present invention, even if the opening characteristic of the discharge amount control valve is non-linear and the non-linear change characteristic is complicatedly changed by the shape change of the valve body or the seating surface of the valve body. The pump characteristic specifying means can accurately learn the pump characteristic representing the relationship between the control amount of the discharge amount control valve and the fuel discharge amount from the fuel pump in the entire fuel discharge region of the fuel pump.

It is a schematic structure figure showing the composition of the whole common rail system of an embodiment. It is explanatory drawing showing the relationship between the I-Qs characteristic of SCV which is a discharge amount control valve, and the TQ-Qp characteristic of PRV which is a pressure-reduction valve. It is a flowchart showing the decompression start point search process performed by ECU. It is a flowchart showing the pump characteristic learning process performed by ECU. It is a flowchart showing the detail of the SCV failure determination process shown in FIG. It is a flowchart showing the pump characteristic specific process shown in FIG.

Embodiments of the present invention will be described below with reference to the drawings.
The present invention is not construed as being limited in any way by the following embodiments. An aspect in which a part of the configuration of the following embodiment is omitted as long as the problem can be solved is also an embodiment of the present invention. Moreover, all the aspects which can be considered in the limit which does not deviate from the essence of the invention specified only by the wording described in the claims are embodiments of the present invention. Further, the reference numerals used in the description of the following embodiments are also used in the claims as appropriate, but this is used for the purpose of facilitating the understanding of the present invention, and limits the technical scope of the present invention. Not intended.

  As shown in FIG. 1, the common rail system 2 of this embodiment is for supplying fuel to, for example, a four-cylinder diesel engine (hereinafter simply referred to as an engine) 4 for automobiles, and stores a high-pressure fuel. 10 is provided.

  The fuel pumped up from the fuel tank 12 is supplied to the common rail 10 in a high-pressure state via the fuel pump 20, and the high-pressure fuel accumulated in the common rail 10 passes through the fuel injection valves 30 provided in each cylinder of the engine 4. Then, the fuel is injected and supplied into the combustion chamber of each cylinder.

  The fuel pump 20 includes a feed pump 22 that pumps fuel from the fuel tank 12 and a fuel pumping unit 24 that discharges the fuel pumped by the feed pump 22 to the common rail 10 at a high pressure.

  In the fuel passage from the feed pump 22 to the fuel pumping unit 24, the fuel suction amount sucked into the fuel pumping unit 24 (in other words, the fuel discharge amount from the fuel pumping unit 24) as a discharge amount control valve of the present invention. ) To adjust the suction control valve (hereinafter referred to as SCV (Suction Control Valve)) 26 is provided.

  The fuel pumping unit 24 pressurizes the fuel sucked through the SCV 26 as the plunger reciprocates as the camshaft cam rotated by the engine 4 rotates, and a delivery valve (not shown) is pressed. And discharged toward the common rail 10.

  Further, the common rail 10 includes a pressure sensor 14 for detecting internal fuel pressure (that is, common rail pressure), and a pressure reducing valve (for reducing the internal fuel pressure by overflowing the internal fuel to the fuel tank 12 side). Hereinafter, PRV (Pressure Reducing Valve)) 16 is provided.

  The PRV 16 is opened by energization and discharges the fuel in the common rail 10 to the fuel tank 12. As shown in FIG. 2, the amount of fuel Qp passing through the pressure reducing valve 16 is determined by the energization time of the pressure reducing valve 16 ( In other words, the valve opening time is proportional to TQ.

  Further, the engine 4 includes a rotation speed sensor 32 for detecting the rotation speed NE, an accelerator sensor 34 for detecting the accelerator operation amount (accelerator opening ACC) by the driver, and a water temperature for detecting the temperature of the cooling water (cooling water temperature THW). A sensor 36, an intake air temperature sensor 38 for detecting the temperature of intake air (intake air temperature TA), and the like are provided.

  The detection signals from these sensors are input to an electronic control unit (hereinafter referred to as ECU) 40 constituted by a microcomputer centering on a CPU, ROM, RAM and the like.

  The ECU 40 receives detection signals from the pressure sensor 14 provided on the common rail 10 and various sensors 32, 34, 36, 38,... Provided on the engine 4, and controls the common rail pressure and the fuel injection from the fuel injection valve 30. Execute control processing.

  The common rail pressure control process is a process for calculating the target pressure of the common rail 10 based on the operating state of the engine 4 and driving and controlling the SCV 26 and the PRV 16 so that the common rail pressure detected by the pressure sensor 14 becomes the target pressure. It is.

  In the fuel injection control process, the fuel injection amount and the fuel injection timing are calculated based on the operating state of the engine 4, and the fuel injection valve 30 of each cylinder is energized for a predetermined energization period at a predetermined timing according to the calculation result. Thus, the fuel injection valve 30 is opened for a predetermined period corresponding to the energization period, and fuel is injected and supplied to each cylinder.

  Next, the ECU 40 learns the pump characteristics of the fuel pump 20 separately from the control processing for such fuel injection control, and based on the learning result, the control amount of the SCV 26 (the energization current I to the SCV 26, , The learning value for correcting the SCV instruction value) is calculated.

  This learning value is used to correct the SCV command value from the design value so that the fuel discharge amount from the fuel pump 20 becomes a desired target amount. , A predetermined discharge ratio (δ%) with respect to the total discharge amount).

  That is, when the ECU 40 determines that the learning condition is satisfied from the travel distance of the vehicle, the elapsed time after the previous learning, etc., the ECU 40 executes the decompression start point search process shown in FIG. 3 and the pump characteristic learning process shown in FIGS. Thus, the current pump characteristic is learned (detected), and the learned value is updated to a value corresponding to the current pump characteristic. Therefore, the ECU 40 functions as the pump characteristic learning device of the present invention.

  Here, the decompression start point search process is a process for detecting the minimum drive amount (decompression start TQ value) necessary for the PRV 16 to open and start the decompression of the common rail pressure as the decompression start point. .

  As shown in FIG. 3, when this process is started, in S110 (S represents a step), the engine 4 is currently in an idle operation state, and this state is used to stabilize the rotational speed NE of the engine 4. It is determined whether or not a predetermined time required has elapsed.

  If the idle operation state has not elapsed for a predetermined time or longer, the process of S110 is executed again to wait for the idle operation state to elapse for a predetermined time or longer. , S120 is entered.

  In S120, the operation mode of the ECU 40 is changed from the normal control mode in which the above-described common rail pressure control process and fuel injection control process are executed to the pressure reduction start point search mode. In the subsequent S130, the current SCV instruction value (that is, during idle operation) is stored in a built-in memory (such as a RAM) as a default value.

Next, in S140, the TQ value, which is the driving amount (energization time) of the PRV 16, is updated to a value obtained by adding a predetermined value T1 (positive value) for searching for a decompression start point to the previous value.
In the common rail pressure control process described above, the common rail pressure is normally controlled to the target pressure (in this case, the target pressure during idle operation) by adjusting the opening of the SCV 26 with the PRV 16 in the closed state. Thus, the initial value of the TQ value before updating in S140 is “0”.

  In the subsequent S150, after updating the TQ value in S140, whether or not the SCV instruction value has changed in a direction to increase the amount of protrusion from the fuel pump 20 by the common rail pressure control process separately performed in the ECU 40. Determine whether.

  That is, when the TQ value is updated in the positive direction in S140, the PRV 16 is driven to the valve opening side. Therefore, when the PRV 16 is actually opened, the common rail pressure decreases from the target pressure, and the SCV command value becomes It changes to the side which increases the fuel discharge amount from the fuel pump 20.

  Therefore, in S150, it is determined whether or not the difference between the latest SCV instruction value set by the common rail pressure control process and the default value stored in S130 exceeds a predetermined value for the valve opening determination of PRV16. Thus, it is determined whether or not the PRV 16 is opened.

  As shown in FIG. 2, the SCV 26 of the present embodiment is configured to increase the fuel discharge amount Qs from the fuel pump 20 as the opening current increases as the energization current I decreases. Therefore, in S150, it is determined whether or not the PRV 16 is opened by determining whether or not the absolute value of the value obtained by subtracting the default value from the SCV instruction value that is the instruction value of the energization current I exceeds a predetermined value. Judging.

  If the difference between the SCV instruction value and the default value does not exceed the predetermined value and it is determined in S150 that the PRV 16 is not opened, the process proceeds to S140 and the PRV 16 TQ value is updated again. To do.

  On the other hand, if the difference between the SCV instruction value and the default value exceeds the predetermined value and it is determined in S150 that the PRV 16 is opened, the process proceeds to S160, and the PRV 16 is subtracted from the TQ value by the predetermined value T2. Is driven in the valve closing direction.

  The predetermined value T2 is set to a positive value smaller than the predetermined value T1 used for updating the TQ value in S140. For this reason, after the valve opening of the PRV 16 is determined in S150, the PRV 16 is driven in the valve closing direction with a smaller change amount than when the valve opening is detected.

  Then, after updating the TQ value in S160, the process proceeds to S170, and it is determined whether or not the difference between the SCV instruction value and the default value is smaller than a predetermined value for valve opening determination of PRV16. , It is determined whether the PRV 16 is closed.

  If it is determined in S170 that the PRV 16 is not closed, the process proceeds to S160 and the TQ value of the PRV 16 is updated again. Conversely, if it is determined that the PRV 16 is closed, the process proceeds to S180. Then, the previous value of the TQ value is stored in a built-in memory (such as a backup RAM or a non-volatile memory) as a TQ value at which the PRV 16 starts reducing the common rail pressure.

In subsequent S190, the operation mode of the ECU 40 is returned from the decompression start point search mode to the normal control mode, and the decompression start point search process is terminated.
In this way, in the decompression start point search process, as shown in FIG. 2, the PRV 16 is gradually driven in the valve opening direction by gradually increasing the TQ value from the initial value “0” by a predetermined value T1.

When the energizing current I of the SCV 26 changes (decreases in this embodiment) by the common rail pressure control, the valve opening of the PRV 16 is detected.
Further, after the valve opening is detected, the TQ value is decreased stepwise by a predetermined value T2 smaller than T1 at the time of valve opening, and the energization current I of the SCV 26 changes (increases in this embodiment) by the common rail pressure control. Then, it is determined that the PRV 16 is opened.

  Then, the previous value of the TQ value at the time of closing the valve is used as the TQ value of the pressure reduction start point at which the PRV 16 opens and the common rail pressure 10 starts to be reduced. Stored in a built-in memory such as a memory.

Next, the pump characteristic learning process is a process executed by the ECU 40 after the learning condition is established and after the pressure reduction start point search process stores the pressure reduction start TQ value.
As shown in FIG. 4, when this process is started, in S210, as in S110 described above, the engine 4 is currently in an idle operation state, and this state is necessary for the rotational speed NE of the engine 4 to be stabilized. It is determined whether or not a predetermined time has elapsed.

  If the idle operation state has not elapsed for a predetermined time or longer, the process of S210 is executed again to wait for the idle operation state to elapse for a predetermined time or longer. , The process proceeds to S220.

  In S220, after changing the operation mode of the ECU 40 from the normal control mode to the learning mode, the process proceeds to S230, and the SCV instruction value at the time of the current idle operation is stored in the built-in memory (backup RAM, nonvolatile memory, etc.). .

  Next, in S240, the SCV instruction value (reference value) at the time of idling operation is read from the reference characteristic representing the SCV instruction value at the time of design, and the deviation amount of the current SCV instruction value from the reference value is calculated. The amount is stored in a built-in memory (such as a backup RAM or a non-volatile memory) as a learning value for a low discharge amount.

  In S250, the amount of change from the learned value at the time of shipment at the low discharge amount obtained in S240 is calculated and stored in a built-in memory (backup RAM, nonvolatile memory, etc.). Note that the learning value at the time of shipment (corresponding to the initial learning value described in the claims) is a learning value that is set in advance to correct a control error caused by variations in the common rail system. It is stored in the ROM together with the characteristics.

Next, in S260, initial values “0” are set for the learning counter and the recording value of the energization time TQ of the PRV 16 (hereinafter referred to as the TQ recording value).
In the subsequent S270, the SCV 26 is opened by a predetermined amount by adding a predetermined update value α (negative value in the present embodiment) to the SCV instruction value, and the amount of fuel discharged from the fuel pump 20 to the common rail 10 ( Increase the fuel discharge amount).

  Next, in S270, when the SCV instruction value is updated in S270, the fuel discharge amount from the fuel pump 20 increases and the common rail pressure rises. Therefore, the process proceeds to S280, and the common rail pressure control is driven to drive the PRV 16. Thus, the system is switched to a method for controlling the common rail pressure to the target pressure.

  In S290, it is determined whether or not the common rail pressure can be maintained at the target pressure by the common rail pressure control switched in S280. If the common rail pressure cannot be maintained at the target pressure, the process proceeds to S310, and the learning counter Is an initial value “0”.

  If it is determined in S310 that the learning counter is not the initial value “0”, in S320, the learning counter and the TQ recording value are cleared to set the initial value “0”, and the process proceeds to S330. . If it is determined in S310 that the learning counter is the initial value “0”, the process proceeds to S330 as it is.

  In S330, the target pressure of the common rail pressure is updated so that the target pressure becomes larger than the previous value by adding a predetermined value β (positive value) to the currently set target pressure (that is, the previous value). .

  In subsequent S340, it is determined whether or not the common rail pressure detected by the pressure sensor 14 (in other words, the actual pressure in the common rail 10) has become smaller than a predetermined value corresponding to the target pressure by updating the target pressure. to decide.

  If it is determined in S340 that the actual pressure is greater than or equal to the predetermined value, it is determined that the fuel in the common rail 10 has not been discharged via the PRV 16, and the PRV 16 is not operating normally, and the process proceeds to S350. Transition.

  In S350, the fact that the PRV 16 has failed is stored in the built-in memory (nonvolatile memory), and the pump characteristic learning process ends. In addition, the process of S290-S350 functions as the 2nd failure determination means of this invention.

  If it is determined in S340 that the actual pressure is smaller than the predetermined value, it is determined that the PRV 16 is operating normally, the process proceeds to S260, and the processes after S260 are executed again.

  Next, if it is determined in S290 that the common rail pressure can be maintained at the target pressure, the process proceeds to S400, and the current TQ value that can maintain the common rail pressure at the target pressure is stored as a TQ recording value in the built-in memory. (Backup RAM, non-volatile memory, etc.)

  In subsequent S410, the learning counter is incremented (+1) to update the count value of the learning counter. In subsequent S420, it is determined whether or not the difference between the TQ value stored this time in S400 and the previous value exceeds a predetermined value for common rail pressure saturation determination.

  If the difference between the TQ value and the previous value exceeds the predetermined value for saturation determination, it is determined that the common rail pressure has increased with the update of the SCV instruction value in S270, and S270 is performed again. Conversely, if the difference between the TQ value and the previous value is less than or equal to the predetermined value, the process proceeds to S430.

  In S430, even if the energization current I to the SCV 26 is changed so that the fuel discharge amount from the fuel pump 20 is increased, the fuel discharge amount does not increase (the saturation point of the IQ characteristic shown in FIG. 2). In order to find the saturation point, the SCV instruction value is updated in the opposite direction to S270.

  That is, in S430, the SCV instruction value is updated by adding an update value γ (positive value in the present embodiment) that is smaller than the update value α in S270 and has a different sign to the SCV instruction value.

  When the SCV instruction value is updated in this way, the SCV 26 is driven in the closing direction and the fuel discharge amount from the fuel pump 20 decreases, so that the process proceeds to S440 and the common rail pressure control is started by the common rail pressure control started in S280. The PRV 16 TQ value necessary for maintaining the pressure is read and stored in the built-in memory (nonvolatile memory) as a TQ recording value.

  In subsequent S450, it is determined whether or not the absolute value of the difference between the TQ value stored this time in S440 and the previous value exceeds a predetermined value for determining the saturation of the common rail pressure. It is determined whether or not the current I has moved through the saturation point shown in FIG. 2 to an adjustable region in which the fuel discharge amount Qs can be changed.

  If it is determined in S450 that the absolute value of the difference between the TQ value and the previous value does not exceed the predetermined value, the process proceeds to S460 again, and the absolute value of the difference between the TQ value and the previous value is the predetermined value. If it is determined that the current has passed, the current I of the SCV 26 is determined to have moved to the adjustable region where the fuel flow rate Q can be changed through the saturation point shown in FIG. Transition.

  In S460, the previous value of the TQ value becomes the drive amount of the PRV 16 necessary for controlling the common rail pressure during idle operation to the target pressure when the fuel discharge amount Qs from the fuel pump 20 becomes maximum. The value is stored in a built-in memory (a backup RAM, a non-volatile memory, etc.) as a TQ value for the total discharge amount.

  In S470, the previous value of the SCV instruction value is used as a final value representing the SCV instruction value at the saturation point at which the fuel discharge amount Qs from the fuel pump 20 is maximum, and the built-in memory (backup RAM, nonvolatile memory, etc.) To remember.

  In S480, the final value of the SCV instruction value is read as a reference value from the reference characteristic representing the SCV instruction value at the time of design, and the deviation amount from the reference value stored in S470 is learned at the time of high discharge amount. The value is stored in a built-in memory (backup RAM, non-volatile memory, etc.).

In the subsequent S490, the amount of change from the learned value at the time of shipment of the learned value at the time of high discharge calculated in S480 is calculated and stored in the built-in memory (backup RAM, nonvolatile memory, etc.).
As described above, in S260 to S470, as shown in FIG. 2, the fuel discharge amount Qs from the fuel pump 20 is changed by changing the SCV instruction value (the energization current I to the SCV 26) when the engine 4 is idling. Increase in stages.

And the common rail pressure which rises with the increase is reduced via PRV16, and a common rail pressure is hold | maintained at the target pressure at the time of idle driving | operation.
Further, by detecting the TQ value (energization time) that is the drive amount of the PRV 16 when the common rail pressure is maintained at the target pressure, the increase in the fuel discharge amount Qs accompanying the change in the SCV instruction value and the PRV 16 are used. A TQ value when the fuel amount Qp discharged from the common rail 10 is balanced is obtained.

Then, the TQ value and the SCV instruction value corresponding to the TQ value are stored in a built-in memory (backup RAM, nonvolatile memory, etc.).
Further, when the SCV instruction value is changed and the fuel discharge amount Qs from the fuel pump 20 is increased, the fuel discharge amount Qs is finally saturated at the maximum discharge amount determined by the plunger volume of the fuel pumping unit 24. Therefore, the saturation point is detected in S430 to S450.

  Then, the SCV instruction value at the saturation point is set as the final value at which the fuel discharge amount Qs from the fuel pump 20 is maximized, and the TQ value at the saturation point is set as the TQ value at the total discharge amount. Stored in a non-volatile memory or the like.

  In S230 to S250, the SCV instruction value at the time of idle operation is stored as the SCV instruction value at the time of the low discharge amount, and the deviation amount of the SCV instruction value from the reference characteristic is used as the learning value at the time of the low discharge amount. Further, the change amount of the learning value from the learning value at the time of shipment is stored.

  Further, in S480 and S490, the deviation amount from the reference characteristic of the SCV instruction value at the time of the high discharge amount is stored as a learning value at the time of the high discharge amount, and the change amount of the learning value from the learning value at the time of shipment is stored. To do.

  In the pump characteristic learning process, in subsequent S500, the failure determination of the SCV 26 is performed based on the learning value at the time of the low discharge amount and the high discharge amount and the change amount from the learning value at the time of shipment of each learning value. An SCV failure determination process is executed.

  In subsequent S600, the decompression start TQ value obtained in the decompression start point search process, the SCV instruction value and TQ value stored each time the SCV instruction value is updated in S270 and S400, and the S460 and S470 are stored. Based on the total discharge amount TQ value and the final value of the SCV command value, a pump characteristic specifying process for specifying the current pump characteristic of the fuel pump 20 is executed.

After executing the pump characteristic specifying process, in S700, the operation mode of the ECU 40 is returned from the learning mode to the normal control mode, and the pump characteristic learning process is ended.
The SCV failure determination process functions as a first failure determination unit of the present invention, and the pump characteristic identification process functions as a pump characteristic identification unit of the present invention.

Next, the SCV failure determination process executed in S500 and the pump characteristic identification process executed in S600 will be described.
As shown in FIG. 5, in the SCV failure determination process, first, in S510, the amount of change from the learning value at the time of low discharge amount from the learning value at the time of shipment, and the learning value at the time of high discharge amount from the learning value at the time of shipment are calculated. It is determined whether or not the amounts of change are both greater than a predetermined value for failure determination.

  If both of these changes are larger than a predetermined value for failure determination, it is determined that the SCV 26 has failed, and the process proceeds to S570, where the determination result (that is, each of the above-mentioned change amounts is both predetermined). Larger than the value) and the failure of the SCV are stored in the built-in memory (nonvolatile memory), and the pump characteristic learning process is terminated. After the completion of the pump characteristics learning, the user is prompted to replace the SCV 26 by notifying the failure of the SCV 26.

  When a negative determination is made in S510, the process proceeds to S520, in which both the high discharge amount learning value and the low discharge amount learning value are both greater than a predetermined failure determination value. Determine whether.

  If these learning values are both greater than a predetermined value for failure determination, it is determined that the SCV 26 has failed, and the process proceeds to S570, where the determination result (that is, both of the learning values are predetermined). Larger than the value) and the failure of the SCV are stored in the built-in memory (nonvolatile memory), and the pump characteristic learning process is terminated.

Next, when a negative determination is made in S520, the process proceeds to S530, in which it is determined whether or not the amount of change from the learning value at the time of shipment at the high discharge amount is larger than a predetermined value for failure determination. To do.
When the amount of change from the learning value at the time of delivery of the high discharge learning value is larger than the predetermined value for failure determination, it is determined that the SCV 26 has failed, and the process proceeds to S570. That is, the change amount of the learning value at the time of high discharge amount is larger than a predetermined value) and the fact that the SCV is faulty are stored in the built-in memory (nonvolatile memory), and the pump characteristic learning process is ended.

Further, when a negative determination is made in S530, the process proceeds to S540, and it is determined whether or not the amount of change from the learning value at the time of shipment of the low discharge amount is larger than a predetermined value for failure determination. .
When the amount of change from the learning value at the time of shipment from the low-learning learning value is larger than a predetermined value for failure determination, it is determined that the SCV 26 has failed, and the process proceeds to S570 and the determination result ( That is, the change amount of the learning value at the time of low discharge amount is larger than the predetermined value) and the fact that the SCV is out of order are stored in the built-in memory (nonvolatile memory), and the pump characteristic learning process is ended.

Next, when a negative determination is made in S540, the process proceeds to S550, and it is determined whether or not the learning value at the time of high discharge amount is larger than a predetermined value for failure determination.
When the high discharge learning value is larger than the predetermined value for failure determination, it is determined that the SCV 26 has failed, and the process proceeds to S570, and the determination result (that is, the high discharge amount learning value is It is stored in the built-in memory (non-volatile memory) that the SCV is faulty, and the pump characteristic learning process is terminated.

When a negative determination is made in S550, the process proceeds to S560, and it is determined whether or not the learning value at the time of low discharge amount is larger than a predetermined value for failure determination.
If the low discharge learning value is greater than the predetermined value for failure determination, it is determined that the SCV 26 has failed, and the process proceeds to S570, where the determination result (that is, the low discharge amount learning value is It is stored in the built-in memory (non-volatile memory) that the SCV is faulty, and the pump characteristic learning process is terminated.

If a negative determination is made in S560, it is determined that the SCV 26 is operating normally, and the process proceeds to the pump characteristic specifying process in S600.
As described above, in the pump characteristic learning process, the failure determination of the SCV 26 is performed based on the learning value at the time of low discharge amount and the learning value at the time of high discharge amount, and the amount of change of each learning value from the learning value at the time of shipment. If the failure is determined, the pump characteristic learning process is terminated and the user is prompted to replace the SCV 26. For this reason, the safety | security at the time of failure of SCV26 can be improved.

  In S570, since the state of each learning value when the failure of the SCV 26 is determined is stored, it is possible to determine how the SCV 26 has deteriorated from the stored information. It can be.

  Next, as shown in FIG. 6, in the pump characteristic specifying process in S600, first in S610, the decompression start TQ value, the total discharge amount TQ value, the final value of the SCV instruction value, and the SCV when searching for the saturation point. The pump characteristics of the fuel pump 20 are calculated based on the SCV instruction value and TQ value stored every time the instruction value is updated, and stored in a built-in memory (nonvolatile memory or the like).

  That is, as shown in FIG. 2, the TQ-Qp characteristic of the PRV 16 is linear, and the fuel amount Qp discharged from the common rail 10 via the PRV 16 is a TQ value from the pressure reduction start TQ value to the TQ value at the time of all discharges. Is proportional to

  Further, the fuel amount Qp when the PRV 16 is driven with the TQ value at the total discharge amount is the fuel amount used for the idle operation of the engine 4 from the fuel discharge amount Qs when the SCV 26 is driven with the final value of the SCV instruction value ( (A certain amount) is reduced.

Therefore, the fuel amount Qp obtained from the TQ value using the TQ-Qp characteristic of the PRV 16 corresponds to the fuel discharge amount Qs from the fuel pump 20.
A plurality of sets of SCV command values and TQ values are stored in the intermediate region from the decompression start point at which the fuel amount Qp discharged from the common rail 10 is “0” to the saturation point at which the fuel amount Qp is maximum. ing.

  Therefore, in S610, based on each of the above information, the TQ value at the total discharge amount is set to 100% of the fuel discharge rate from the fuel pump 20, and the SCV instruction value (that is, the energization current I of the SCV 26) The pump characteristic (that is, the I-Qs characteristic shown in FIG. 2) representing the relationship with the discharge ratio (and hence the fuel discharge amount Qs) is calculated.

  Next, in S620, using the I-Qs characteristic obtained in S610, a preset predetermined ratio (δ%) learning point with respect to the total discharge amount from the fuel pump 20 (fuel discharge ratio 100%). The SCV instruction value at is calculated.

  In subsequent S630, the deviation amount from the reference characteristic of the SCV instruction value at the learning point calculated in S620 is stored in a built-in memory (non-volatile memory or the like) as a learned value at the fuel discharge amount δ%.

  Next, in S640, it is determined whether or not learning values have been calculated at all the learning points set in advance. If calculation of learning values at all learning points has not been completed, the learning value is determined in S650. The ratio (δ%) of learning points to be calculated is updated, and the process proceeds to S620 again.

If it is determined in S640 that calculation of learning values at all learning points has been completed, the pump characteristic specifying process is terminated, and the process proceeds to S700.
As described above, in the common rail system of the present embodiment, the ECU 40 for fuel injection control functions as the pump characteristic learning device of the present invention, and the engine 4 is operated at a constant speed with no load, The pump characteristic of the fuel pump 20 is learned (detected), and the learning value for correcting the SCV instruction value used in the common rail pressure control is updated.

  The learning (detection) of the pump characteristics is performed not only at the two points of the point at which the fuel discharge amount from the fuel pump 20 is maximized and the point at which it is minimized, but also at an intermediate point between these two points. A pump characteristic (I-Qs characteristic) representing the relationship between the SCV instruction value and the fuel discharge amount Qs is obtained in the entire fuel discharge region of the fuel pump 20.

  For this reason, it becomes possible to set a learning value that can appropriately correct the SCV instruction value in accordance with the change regardless of how the pump characteristic changes due to the change over time. It is possible to control so that the fuel injection characteristic to the engine 4 is always optimum.

  Further, in the present embodiment, when the fuel discharge amount from the fuel pump 20 to the common rail 10 is increased via the SCV 26, if the common rail pressure cannot be controlled to the target pressure via the PRV 16, a failure determination of the PRV 16 is performed.

  This failure determination is performed by changing the target pressure of the common rail 10. If the common rail pressure cannot be changed by the PRV 16 even if the target pressure is changed, it is determined that the PRV 16 has failed.

  Therefore, according to the common rail system of the present embodiment, the failure determination of the PRV 16 can be performed when learning the pump characteristics, so that the reliability of the common rail system can be improved.

  Further, in the present embodiment, after the learning value at the time of low discharge amount and the learning value at the time of high discharge amount are calculated, SCV failure determination processing is executed to perform not only PRV 16 failure determination but also SCV 26 failure determination.

In the SCV failure determination process, when the failure of the SCV 26 is determined, the determination result is stored so that it is possible to grasp how the SCV 26 has deteriorated.
Therefore, in the present embodiment, the reliability of the common rail system can be further improved, and the designer can easily identify the cause of the failure (deterioration), thereby improving the durability of the SCV 26 by improving the SCV 26. Can also be increased.

As mentioned above, although one Embodiment of this invention was described, this invention is not limited to the said embodiment, In the range which does not deviate from the summary of this invention, a various aspect can be taken.
For example, in the above embodiment, in the pump characteristic specifying process, I-Qs in the fuel pump 20 is calculated as the pump characteristic, and the learning value is calculated from the pump characteristic.

  On the other hand, in the pump characteristic specifying process, the pump characteristic is simply obtained, and the learning value (SCV instruction value correction value) is calculated in the common rail pressure control process performed in the normal control mode. Good.

  Further, the learning of the pump characteristics is only required when the pump characteristics deviate from the time of design or shipment and the fuel discharge amount from the fuel pump 20 can be controlled by appropriately correcting the SCV instruction value according to the deviation. .

  For this reason, it is not always necessary to calculate the pump characteristic in the pump characteristic specifying process, and the fuel discharge ratio δ% is obtained by proportional calculation using the TQ-Qp characteristic obtained from the depressurization start point TQ value and the total discharge time TQ value. The learning value may be updated by obtaining the SCV instruction value at.

  2 ... Common rail system, 4 ... Engine, 10 ... Common rail, 12 ... Fuel tank, 14 ... Pressure sensor, 16 ... PRV (pressure reducing valve), 20 ... Fuel pump, 22 ... Feed pump, 24 ... Fuel pumping section, 26 ... SCV (Intake regulating valve), 30 ... fuel injection valve, 32 ... rotational speed sensor, 34 ... accelerator sensor, 36 ... water temperature sensor, 38 ... intake air temperature sensor, 40 ... ECU (electronic control unit).

Claims (7)

A common rail (10) for accumulating high-pressure fuel to be supplied to the fuel injection valve (30) of the engine (4);
A fuel pump (20) driven by the engine to supply high pressure fuel to the common rail;
A discharge amount control valve (26, 60) provided in the fuel pump for adjusting a fuel discharge amount from the fuel pump to the common rail;
A pressure reducing valve (16) having a driving characteristic in which the fuel pressure in the common rail is reduced by discharging the fuel from the common rail, and the relationship between the fuel discharging amount and the driving amount is linear;
In a common rail system comprising: a fuel that learns a pump characteristic representing a relationship between a control amount of the discharge amount control valve and a fuel discharge amount from the fuel pump when the engine is operated at a constant speed. A pump characteristic learning device for a pump,
A change in which the control amount of the discharge amount control valve changes by gradually opening the pressure reducing valve from the closed state while controlling the fuel pressure in the common rail to the target pressure via the discharge amount control valve. Pressure reduction start point search means (40, S110 to S180) for searching for a point and storing the drive amount of the pressure reducing valve at the change point as a pressure reduction start drive amount;
The fuel pressure in the common rail is reduced by sequentially increasing the fuel discharge amount from the fuel pump to the common rail via the discharge amount control valve and discharging the fuel in the common rail via the pressure reducing valve. A discharge amount increase control means (40, S270 to S420) for controlling to a target pressure;
Control that stores the control amount of the discharge amount control valve and the drive amount of the pressure reducing valve required to control the fuel pressure to the target pressure each time the discharge amount increase control means increases the fuel discharge amount Amount storage means (40, S400);
If the drive amount of the pressure reducing valve does not change even when the discharge amount increase control means increases the fuel discharge amount, it is determined that the discharge amount control valve has entered a saturation region where the fuel discharge amount cannot be increased, By terminating the operation of the discharge amount increase control means and gradually changing the control amount of the discharge amount control valve by a change amount smaller than the change amount at the time of increase by the discharge amount increase control means, A saturation point that is a boundary between the discharge amount controllable region where the drive amount changes and the saturation region is searched, and the drive amount of the pressure reducing valve and the control amount of the discharge amount control valve at the saturation point are all Saturation point searching means (40, S430 to S470) for storing as a discharge amount driving amount and a final control amount,
The decompression start drive amount stored in the decompression start point search means, the total discharge amount drive amount and final control amount stored in the saturation point search means, and the fuel discharge amount in the control amount storage means Pump characteristic specifying means (40, S240, S480, S600, S610 to S650) for specifying the pump characteristic based on the control amount and the drive amount stored for each increase amount;
With
The pump characteristic specifying means is based on a linear characteristic representing the relationship between the drive amount of the pressure reducing valve and the fuel discharge amount from the fuel pump, obtained from the pressure reduction start drive amount and the total discharge amount drive amount. A pump characteristic learning device for a fuel pump, wherein a pump characteristic representing a relationship between a control amount of the discharge amount control valve and a fuel discharge amount from the fuel pump is specified in an entire fuel discharge region of the fuel pump.   The pump characteristic specifying means is based on a linear characteristic representing the relationship between the drive amount of the pressure reducing valve and the fuel discharge amount from the fuel pump, obtained from the pressure reduction start drive amount and the total discharge amount drive amount. The control amount of the discharge amount control valve required to set the fuel discharge amount from the fuel pump to a predetermined intermediate discharge amount is obtained, and the deviation from the reference value of the control amount is set as a learning value representing the pump characteristics The pump characteristic learning device for a fuel pump according to claim 1, wherein:   The pump characteristic specifying means includes an amount of deviation from a reference value of the final control amount, and the discharge amount control valve before the discharge amount increase control means increases the fuel discharge amount via the discharge amount control valve. 3. The pump characteristic learning device for a fuel pump according to claim 2, wherein at least one of the deviation amount of the control amount from the reference value is set as a learning value at the time of high discharge amount or a learning value at the time of low discharge amount. . It is determined whether the learning value set by the pump characteristic specifying means or the amount of change of the learning value from the initial learning value is within a preset allowable range, and is not within the allowable range. A first failure determination means (40, S500, S510 to S570) for determining a failure of the discharge amount control valve,
The pump characteristic learning device for a fuel pump according to any one of claims 1 to 3, further comprising: Even if the discharge amount increase control means sequentially increases the fuel discharge amount from the fuel pump to the common rail via the discharge amount control valve, the fuel in the common rail is discharged via the pressure reducing valve, Second failure determination means (40, S290 to S350) for determining failure of the pressure reducing valve when the fuel pressure in the common rail is not controlled to a target pressure;
The pump characteristic learning device for a fuel pump according to any one of claims 1 to 4, further comprising:   The second failure determination means changes the target pressure when the fuel pressure in the common rail is not controlled to the target pressure, and the pressure in the common rail does not change even if the target pressure is changed. 6. The pump characteristic learning apparatus for a fuel pump according to claim 5, wherein the pressure reducing valve is determined to have failed.   The pump of the fuel pump according to any one of claims 1 to 6, wherein the pump characteristic learning device operates when the engine is in an idle operation state and learns the pump characteristic. Characteristic learning device.

Images