CONTROL DEVICE FOR DIESEL ENGINE

Abstract

When a diesel engine is determined to be in a motoring state, a hysteresis zero angle H.sub.0 is determined (step S14). Subsequently, a gradient d.sub.n is calculated (step S16). The gradient d.sub.n is calculated based on data (θ.sub.n, Δh.sub.n) of a deviation Δh.sub.n at a retardation side from the hysteresis zero angle H.sub.0 and at an advance side from a predetermined crank angle. Subsequently, the gradient d.sub.n and the hysteresis zero angle H.sub.0 are updated (step S18). When the diesel engine is determined to be in a non-motoring state, data (θ.sub.n, P.sub.n) of an actual in-cylinder pressure is corrected based on a newest correction coefficient η and hysteresis zero angle H.sub.0 (step S22).

Claims

1. A control device for a diesel engine which is configured to control a diesel engine including a glow plug integrated type in-cylinder pressure sensor, wherein the control device is configured to: determine a crank angle at which a difference, which is obtained by subtracting a reference in-cylinder pressure at a crank angle at which an actual in-cylinder pressure is detected, from the actual in-cylinder pressure detected by the in-cylinder pressure sensor at each of predetermined crank angles in a cycle in which the diesel engine is in a motoring state, changes from negative to positive; calculate a rate of change of a pressure difference between the actual in-cylinder pressure and the reference in-cylinder pressure in a predetermined crank angle region at a retardation side from the crank angle at which the difference changes from negative to positive; and correct the actual in-cylinder pressure detected at each of the predetermined crank angles in the cycle in which the diesel engine is in the non-motoring state, based on a crank angle interval from the crank angle at which the actual in-cylinder pressure is detected to the crank angle at which the difference changes from negative to positive, and the rate of change.

2. The control device for a diesel engine according to claim 1, wherein the crank angle at which the difference changes from negative to positive is a crank angle at a retardation side from a compression top dead center.

3. The control device for a diesel engine according to claim 1, wherein the control device is further configured to correct the rate of change by relating the rate of change to an amount of fuel that is injected into a cylinder of the diesel engine, wherein the rate of change is corrected to a smaller value as the amount of fuel is larger.

4. The control device for a diesel engine according to claim 3, wherein the control device is further configured to correct the rate of change by relating the rate of change to an engine speed in the motoring state.

5. The control device for a diesel engine according to claim 1, wherein the control device is further configured to before calculating the difference: correct data of the actual in-cylinder pressure so that a maximum value of the actual in-cylinder pressure detected by the in-cylinder pressure sensor at each of the predetermined crank angles in the cycle in which the diesel engine is in the motoring state becomes equal to a maximum value of the reference in-cylinder pressure; and correct the data of the actual in-cylinder pressure so that a crank angle showing the maximum value corresponds to a top dead center.

6. The control device for a diesel engine according to claim 1, wherein the control device is further configured to correct the actual in-cylinder pressure detected at each of the predetermined crank angles in the cycle in which the diesel engine is in the non-motoring state, based on equation (1) as follows:
correction value of actual in-cylinder pressure=value P.sub.n of actual in-cylinder pressure×(detection crank angle θ.sub.n−hysteresis zero angle H.sub.0)×correction coefficient η   (1), in equation (1), the value P.sub.n of the actual in-cylinder pressure means the actual in-cylinder pressure detected at each of the predetermined crank angles in the cycle in which the diesel engine is in the non-motoring state, the detection crank angle θ.sub.n means the crank angle at which the value P.sub.n of the actual in-cylinder pressure is detected, the hysteresis zero angle H.sub.0 means the crank angle at which the difference changes from negative to positive, and the correction coefficient η means the rate of change.

7. The control device for a diesel engine according to claim 1, wherein the control device is further configured to correct the actual in-cylinder pressure detected at each of the predetermined crank angles in the cycle in which the diesel engine is in the non-motoring state, based on equation (2) as follows:
correction value P.sub.n of actual in-cylinder pressure=value P.sub.n of actual in-cylinder pressure×(detection crank angle θ.sub.n−hysteresis zero angle H.sub.0)×correction coefficient η/100  (2), in equation (2), the value P.sub.n of the actual in-cylinder pressure means the actual in-cylinder pressure detected at each of the predetermined crank angles in the cycle in which the diesel engine is in the non-motoring state, the detection crank angle θ.sub.n means the crank angle at which the value P.sub.n of the actual in-cylinder pressure is detected, the hysteresis zero angle H.sub.0 means the crank angle at which the difference changes from negative to positive, and the correction coefficient η means a percentage of a value obtained by dividing the rate of change by the reference in-cylinder pressure.

Description

BRIEF DESCRIPTION OF DRAWINGS

[0034] FIG. 1 is a schematic diagram explaining a configuration of a diesel engine to which a control device according to a first embodiment of the present application is applied;

[0035] FIG. 2 is a functional block diagram of an ECU 20 as the control device according to the first embodiment of the present application;

[0036] FIG. 3 is a diagram explaining a problem in an in-cylinder pressure sensor 10;

[0037] FIG. 4 is a diagram explaining a problem in the in-cylinder pressure sensor 10;

[0038] FIG. 5 is a diagram illustrating processing images of data of actual in-cylinder pressures in a P.sub.max correction section 46a and a TDC correction section 46b illustrated in FIG. 2;

[0039] FIG. 6 is a diagram explaining hysteresis between a correction value P.sub.n′ of an actual in-cylinder pressure in a motoring state, and a value PR.sub.n of a reference in-cylinder pressure;

[0040] FIG. 7 is a diagram explaining details of the hysteresis between “reference values” and “sensor values” observed in FIG. 4 and FIG. 6;

[0041] FIG. 8 is a diagram explaining details of the hysteresis between the “reference values” and the “sensor values” observed in FIG. 4 and FIG. 6;

[0042] FIG. 9 is a diagram illustrating a processing image of a deviation Δh.sub.n in a hysteresis zero angle determination section 46c and a gradient calculation section 46d illustrated in FIG. 2:

[0043] FIG. 10 is a diagram illustrating a processing image of data of an actual in-cylinder pressure in an in-cylinder pressure correction section 44 illustrated in FIG. 2;

[0044] FIG. 11 is a flowchart illustrating one example of data processing of an actual in-cylinder pressure realized by a function of the ECU 20;

[0045] FIG. 12 is a functional block diagram of the ECU 20 as a control device according to a second embodiment of the present application:

[0046] FIG. 13 is a diagram explaining a method for relating a gradient d.sub.n to a fuel injection amount; and

[0047] FIG. 14 is one example of a diagram illustrating the relationship illustrated in FIG. 13 at each engine speed in the motoring state.

DESCRIPTION OF EMBODIMENTS

[0048] Hereunder, one or more embodiments of the present application will be described based on the drawings. Note that common elements in the respective drawings are assigned with the same reference signs, and redundant explanation will be omitted. Further, the present application is not limited by the one or more embodiments as follows.

First Embodiment

[0049] First of all, a first embodiment of the present application will be described with reference to FIG. 1 to FIG. 11.

[0050] FIG. 1 is a schematic diagram explaining a configuration of a diesel engine to which a control device according to the first embodiment of the present application is applied. The diesel engine to which the control device according to the present embodiment is applied includes a glow plug integrated type in-cylinder pressure sensor 10. The in-cylinder pressure sensor 10 includes a hollow housing 12. In a shaft hole of the housing 12, a pressure receiving section 14 that also functions as a heater rod is inserted. The pressure receiving section 14 is configured to be movable in an axial direction thereof. When a pressure in a cylinder changes, a magnitude of a load that is transmitted to a sensing section 16 via the pressure receiving section 14 changes. The change of the load is detected by the sensing section 16, and thereby the pressure in the cylinder is detected. A seal member 18 is provided between the housing 12 and the pressure receiving section 14. The seal member 18 is provided, and thereby leakage of gas in the cylinder to an outside is prevented.

[0051] The control device according to the present embodiment is realized as a part of a function of the ECU 20 that controls the diesel engine. The ECU 20 takes in and processes signals of various sensors installed in the diesel engine. FIG. 2 illustrates various sensors that are connected to the ECU 20. As illustrated in FIG. 2, the various sensors include at least an accelerator opening degree sensor 22, a crank angle sensor 24, a water temperature sensor 26 and an ignition switch 28, besides the in-cylinder pressure sensor 10 illustrated in FIG. 1.

[0052] The accelerator opening degree sensor 22 detects a depressing amount on an accelerator pedal. The crank angle sensor 24 detects a rotation angle of a crankshaft. The water temperature sensor 26 detects a cooling water temperature of the diesel engine. The ignition switch 28 receives an instruction to supply/stop electric power to an electric power system of the diesel engine.

[0053] The ECU 20 processes the signals taken in from the aforementioned various sensors, and operates various actuators of a system in accordance with a predetermined control program. FIG. 2 also illustrates various actuators that are connected to the ECU 20. As illustrated in FIG. 2, the various actuators include at least an injector 30 that injects fuel into the cylinder of the diesel engine, besides a heater rod 10a of the in-cylinder pressure sensor 10.

[0054] FIG. 2 illustrates functional blocks of the ECU 20 as the control device according to the present embodiment. As illustrated in FIG. 2, the ECU 20 includes an operation state determination section 32, a glow plug drive control section 34, a fuel injection amount setting section 36, a fuel injection timing setting section 40 and an injector drive control section 38.

[0055] The operation state determination section 32 determines an operation state of the diesel engine. The glow plug drive control section 34 performs energization (glow energization) to the heater rod 10a while a cooling water temperature that is detected by the water temperature sensor 26 is lower than a predetermined temperature at a time of a cold start of the diesel engine. The energization is started by an ON signal from the ignition switch 28. Further, the energization is ended at a time point at which the cooling water temperature rises to a temperature higher than a predetermined temperature. The fuel injection amount setting section 36 sets a fuel injection amount corresponding to a depressing amount on the accelerator pedal based on the operation state determined by the operation state determination section 32. The fuel injection amount setting section 36 also outputs the set fuel injection amount to the injector drive control section 38. The fuel injection timing setting section 40 sets an injection mode, an injection timing and the like that correspond to the fuel injection amount which is set by the fuel injection amount setting section 36. The fuel injection timing setting section 40 also outputs the injection mode, the injection timing and the like which are set, to the injector drive control section 38.

[0056] The ECU 20 also includes a storage section 42, an in-cylinder pressure correction section 44, and a correction coefficient setting section 46, as components for performing correction of the in-cylinder pressure detected by the in-cylinder pressure sensor 10.

[0057] The storage section 42 stores a correction coefficient η for correcting a value P.sub.n of the actual in-cylinder pressure, and a value PR.sub.n of a reference in-cylinder pressure. The value P.sub.n of the actual in-cylinder pressure is a value of an in-cylinder pressure which is detected at each predetermined crank angle θ.sub.1 by the in-cylinder pressure sensor 10. The value PR.sub.n of the reference in-cylinder pressure is a value (an initial value) of an in-cylinder pressure at a time of the sensor being brand new, and in a motoring state. As the initial value, a value of an in-cylinder pressure that is obtained in advance by an in-cylinder pressure sensor having a configuration equivalent to the configuration of the in-cylinder pressure sensor 10 is usually used.

[0058] Here, the motoring state refers to an operation state in which the engine speed of the diesel engine is in a low speed region, the depressing amount on the accelerator pedal is zero and fuel is not injected from the injector 30. A region of 3000 rpm or less is cited as an example of the low speed region, but it is needless to say that the low speed region should be properly changed in accordance with an engine system. The correction coefficient η is updated each time the correction coefficient η is set in the correction coefficient setting section 46. Note that details of the correction coefficient η will be described later. Further, the storage section 42 is assumed to store data necessary to realize the function in the present embodiment in addition to the correction coefficient η.

[0059] Before explaining the in-cylinder pressure correction section 44 and the correction coefficient setting section 46, a problem in the in-cylinder pressure sensor 10 will be described with reference to FIGS. 3 and 4. As illustrated in FIG. 3, a small gap exists between a sensor insertion hole 48 that is formed in the cylinder head, and the pressure receiving section 14. When soot that is generated in the cylinder, unburned fuel, engine oil and the like enter the gap, and adhere to a wall surface of the sensor insertion hole 48 and a surface of the pressure receiving section 14, the soot, unburned fuel, engine oil and the like may become deposits. When soot and the like further adhere to the deposits, or the deposits bond to deposits around them, the deposits accumulate. When deposits adhere to or accumulate on the surface of the pressure receiving section 14, a movement characteristic in an axial direction of the pressure receiving section 14 changes.

[0060] When the movement characteristic in the axial direction of the pressure receiving section 14 changes, hysteresis occurs to the value P.sub.n of the actual in-cylinder pressure. FIG. 4 is a diagram explaining the hysteresis. FIG. 4 is created when fuel is injected from the injector 30, that is, when the diesel engine is in a non-motoring state. In FIG. 4, the value P.sub.n of the actual in-cylinder pressure at a time of deposit adherence/accumulation is expressed by a “sensor value”, and the value P.sub.n of the actual in-cylinder pressure at a time of a sensor being brand new is expressed by a “reference value”. As is understandable when tendencies of the “sensor value” and the “reference value” illustrated in FIG. 4 are compared, hysteresis is observed between both of them. Describing in more detail, from 60° before a compression top dead center to a vicinity of 10° after the compression top dead center, the “reference value” is substantially larger than the “sensor value”. Contrary to the above, at a retardation side from the vicinity of 10° after the compression top dead center, the “reference value” is substantially smaller than the “sensor value”.

[0061] Returning to FIG. 2, explanation of the functions of the ECU 20 will be continued. The correction coefficient setting section 46 sets the correction coefficient η by using data of the reference in-cylinder pressure received from the storage section 42, and data of the actual in-cylinder pressure in the motoring state. As components for setting the correction coefficient η, the correction coefficient setting section 46 includes a P.sub.max correction section 46a, a TDC correction section 46b, a hysteresis zero angle determination section 46c and a gradient calculation section 46d.

[0062] First, the P.sub.max correction section 46a and the TDC correction section 46b will be described. In the P.sub.max correction section 46a and the TDC correction section 46b, preprocessing of the data of the actual in-cylinder pressure in the motoring state is performed. As already described, the actual in-cylinder pressure is detected at each of predetermined crank angles θ.sub.1 (intervals of 10° as an example). Consequently, when the values P.sub.n of the actual in-cylinder pressure from 60° before the compression top dead center to 60° after the compression top dead center are arranged in sequence of the detection crank angles, data (θ.sub.n, P.sub.n) is expressed by (θ.sub.BTDC60°, P.sub.BTDC60°), (θ.sub.BTDC60°+θ1, P.sub.BTDC60°+θ1), . . . , (θ.sub.ATDC60°−θ1, P.sub.ATDC60°−θ1), and (θ.sub.ATDC60°, P.sub.ATDC60°).

[0063] The P.sub.max correction section 46a corrects the data (θ.sub.n, P.sub.n) so that a maximum value P.sub.n.sub._.sub.max of the actual in-cylinder pressure in the motoring state becomes equal to a maximum value PR.sub.n.sub.—max of the reference in-cylinder pressure. More specifically, the P.sub.max correction section 46a firstly detects a pressure difference ΔP.sub.n.sub._.sub.max between the maximum value P.sub.n.sub.—max and the maximum value PR.sub.n.sub._.sub.max. When the pressure difference ΔP.sub.n.sub._.sub.max is detected, the P.sub.max correction section 46a corrects the value P.sub.n of the actual in-cylinder pressure to increase the value P.sub.n by the pressure difference ΔP.sub.n.sub._.sub.max, or decrease the value P.sub.n by ΔP.sub.n.sub._.sub.max.

[0064] The TDC correction section 46b corrects the data (θ.sub.n, P.sub.n) so that a crank angle showing the maximum value P.sub.n.sub._.sub.max of the actual in-cylinder pressure in the motoring state corresponds to a TDC. More specifically, the TDC correction section 46b firstly detects a phase difference Δθ of the crank angle showing the maximum value P.sub.n.sub._.sub.max and the TDC. When the phase difference Δθ is detected, the TDC correction section 46b corrects the value of a detection crank angle θ.sub.n to a retardation side or an advance side by the phase difference Δθ.

[0065] FIG. 5 illustrates a processing image of the data (θ.sub.n, P.sub.n) in the P.sub.max correction section 46a and the TDC correction section 46b. In FIG. 5, the maximum value PR.sub.n.sub._.sub.max is larger than the maximum value P.sub.n.sub._.sub.max. Consequently, in the P.sub.max correction section 46a, the value P.sub.n of the actual in-cylinder pressure is corrected to increase by an amount of the pressure difference ΔP.sub.n.sub._.sub.max. Further, in FIG. 5, the crank angle showing the maximum value P.sub.n.sub._.sub.max is located at a retardation side from the TDC. Consequently, in the TDC correction section 46b, the value of the detection crank angle θ.sub.n is corrected to the advance side by the amount of the phase difference Δθ.

[0066] Next, the hysteresis zero angle determination section 46c and the gradient calculation section 46d will be described. The hysteresis zero angle determination section 46c calculates a deviation Δh.sub.h of a correction value P.sub.n′ of the actual in-cylinder pressure to the value of the reference in-cylinder pressure PR.sub.n based on correction data (θ.sub.n′, P.sub.n′) of the actual in-cylinder pressure, and data (θ.sub.n, PR.sub.n) of the values of the reference in-cylinder pressure PR.sub.n arranged in sequence of the detection crank angles. The hysteresis zero angle determination section 46c determines a crank angle (hereinafter, also referred to as “a hysteresis zero angle H.sub.0”) at which a value of the deviation Δh.sub.n changes from negative to positive. The gradient calculation section 46d calculates a gradient d.sub.n of the deviation Δh.sub.n in a crank angle region at a retardation side from the hysteresis zero angle H.sub.0.

[0067] FIG. 6 is a diagram explaining hysteresis between the correction value P.sub.n′ of the actual in-cylinder pressure in the motoring state, and the PR.sub.n of the reference in-cylinder pressure. In FIG. 6, the correction value P.sub.n′ of the actual in-cylinder pressure at a time of deposit adherence/accumulation is expressed by a “sensor value” and the value P.sub.n of the actual in-cylinder pressure at the time of the sensor being brand new is expressed by a “reference value”. As is understandable when tendencies of the “sensor value” and the “reference value” illustrated in FIG. 6 are compared, hysteresis is observed between both of them. Describing in more detail, the “reference value” becomes lower than the “sensor value” in a retardation side from a vicinity of 10° after the compression top dead center. However, as is understandable when FIG. 6 is compared with FIG. 4, FIG. 6 illustrating the tendency in the motoring state, and FIG. 4 illustrating the tendency in the non-motoring state differ in the crank angle region where hysteresis is observed.

[0068] FIGS. 7 and 8 are diagrams explaining details of the hysteresis of the “reference values” and the “sensor values” observed in FIGS. 4 and 6. The same diagram as in FIG. 4 is illustrated on an upper tier of FIG. 7, and the deviation Δh.sub.n of the “sensor value” to the “reference value” illustrated in FIG. 4 is illustrated in a lower tier in FIG. 7. Further, the same diagram as in FIG. 6 is drawn on an upper tier in FIG. 8, and the deviation Δh.sub.n of the “sensor value” to the “reference value” illustrated in FIG. 6 is illustrated in a lower tier in FIG. 8. The deviation Δh.sub.n of the “sensor value” to the “reference value” is expressed as a percentage calculated by substituting the “reference value” and the “sensor value” at the same crank angle into equation (1) as follows.

Deviation Δh.sub.n [%]=100×(sensor value−reference value)/reference value  (1)

[0069] In the lower tier in FIG. 7, a gradient (that is, a rate of change of the deviation Δh.sub.n) of the deviation Δh.sub.n is substantially constant. In the lower tier in FIG. 8, a gradient of the deviation Δh.sub.n becomes small in a crank region at an advance side from the vicinity of 10° after the compression top dead center. In particular, in the crank angle region from the a vicinity of 30° before the compression top dead center to the vicinity of 10° after the compression top dead center, the gradient of the deviation Δh.sub.n becomes remarkably small. The cause of occurrence of such a difference is not certain. However, the present inventor surmises that the difference between the maximum value P.sub.n.sub._.sub.max and the minimum value P.sub.n.sub._.sub.min of the actual in-cylinder pressure is originally small because combustion is not performed in the motoring state, and therefore hysteresis is difficult to observe in the crank angle region at the advance side from the compression top dead center.

[0070] The hysteresis zero angle determination section 46c and the gradient calculation section 46d perform processing in which characteristics peculiar to the motoring state like this are taken into consideration. FIG. 9 is a diagram illustrating a processing image of the deviation Δh.sub.n in the hysteresis zero angle determination section 46c and the gradient calculation section 46d. FIG. 9 illustrates the deviation Δh.sub.n illustrated in the lower tier in FIG. 8, which is in the vicinity of 10° after the compression top dead center. The hysteresis zero angle H.sub.0 is determined as a crank angle at a time of the value of the deviation Δh.sub.n illustrated by the solid line corresponds to zero. Further, from the value of the deviation Δh.sub.n in the crank angle region at the retardation side from the hysteresis zero angle H.sub.0, a gradient d.sub.n illustrated by a broken line is calculated.

[0071] In calculation of the gradient d.sub.n, data (θ.sub.n, Δh.sub.n) of the deviation Δh.sub.n at the retardation side from the hysteresis zero angle H.sub.0, and at an advance side from a predetermined crank angle (60° after the compression top dead center, as an example) is used. In other words, data (θ.sub.n, Δh.sub.n) of the deviation Δh.sub.n at an advance side from the hysteresis zero angle H.sub.0, and data (θ.sub.n, Δh.sub.n) at a retardation side from the predetermined crank angle are not used in calculation of the gradient d.sub.n.

[0072] Calculation of the gradient d.sub.n is performed by using data (θ.sub.n, Δh.sub.n) of at least two deviations Δh.sub.n.

[0073] For example, when arbitrary two data (θ.sub.k+θ1, Δh.sub.k+θ1) and (θ.sub.k−θ1, Δh.sub.k−θ1) of the data (θ.sub.n, Δh.sub.n) of the deviation Δh.sub.n in the aforementioned crank angle region are used, the gradient d.sub.n can be obtained by equation (4) as follows.

Gradient d.sub.n=gradient d.sub.k=(Δh.sub.k+1−Δh.sub.k−1)/(θ.sub.k+1−θ.sub.k−1)  (4)

[0074] Further, for example, when all the data (θ.sub.n, Δh.sub.n) of the deviation Δh.sub.n in the aforementioned crank angle region are used, the gradient d.sub.n can be obtained by equation (5) as follows.

Gradient d.sub.n=average of gradient d.sub.k=ave{(Δh.sub.n+θ1−Δh.sub.n−θ1)/(θ.sub.n+θ1−θ.sub.n−θ1)}  (5)

[0075] Further, for example, a plurality of gradients d.sub.k obtained by equation (4) described above are obtained within the aforementioned crank angle region, and a maximum value of the plurality of gradients d.sub.k can be used as the gradient d.sub.n (refer to equation (6) as follows).

Gradient d.sub.n=maximum value of gradient d.sub.k=max{(Δh.sub.n+θ1−Δh.sub.n−θ1)/(θ.sub.n+θ1−θ.sub.n−θ1)}  (6)

[0076] The correction coefficient setting section 46 outputs the value of the hysteresis zero angle H.sub.0 to the storage section 42. In addition, the correction coefficient setting section 46 outputs the value of the gradient d.sub.n to the storage section 42 as the correction coefficient η.

[0077] The in-cylinder pressure correction section 44 receives the correction coefficient η and the date of the hysteresis zero angle H.sub.0 from the storage section 42, and performs correction of the value P.sub.n of the actual in-cylinder pressure in the non-motoring state. The value P.sub.n of the actual in-cylinder pressure that configures the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure in the non-motoring state is corrected by equation (7) as follows that uses a crank angle interval from the hysteresis zero angle H.sub.0 to the detection crank angle θ.sub.n for example, and the correction coefficient η.

Correction value P.sub.n of actual in-cylinder pressure=value P.sub.n of actual in-cylinder pressure×(detection crank angle θ.sub.n−hysteresis zero angle H.sub.0)×correction coefficient η/100  (7)

[0078] FIG. 10 is a diagram illustrating a processing image of data of the actual in-cylinder pressure in the in-cylinder pressure correction section 44. As illustrated in FIG. 10, in the in-cylinder pressure correction section 44, the value P.sub.n of the actual in-cylinder pressure is corrected to an increase side, in the crank angle region at the advance side from the compression top dead center, when classifying roughly. Contrary to the above, in the crank angle region at the retardation side from the compression top dead center, the value P.sub.n of the actual in-cylinder pressure is corrected to a decrease side. That is, the data of the actual in-cylinder pressure is corrected in a direction to eliminate the hysteresis of the “sensor value” and the “reference value” explained with FIG. 4.

[0079] FIG. 11 is a flowchart illustrating an example of data processing of the actual in-cylinder pressure which is realized by the function of the ECU 20 described above. Note that the routine illustrated in FIG. 11 is repeatedly executed at each cycle after start of the diesel engine.

[0080] In the routine illustrated in FIG. 11, it is firstly determined whether the diesel engine is in a motoring state (step S10). In the present step, processing as follows is performed. First, an engine speed is calculated based on a detection value of the crank angle sensor 24. Further, a detection value of the accelerator opening degree sensor 22 is acquired. When the engine speed is less than a threshold value, and the detection value of the accelerator opening degree sensor 22 is equal to zero, it is determined that the diesel engine is in the motoring state. When the engine speed is the threshold value or more, or the detection value of the accelerator opening degree sensor 22 is not zero, it is determined that the diesel engine is in the non-motoring state.

[0081] When it is determined that the diesel engine is in the motoring state in step S10, the preprocessing of the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure is performed (step S12). In the present step, processing as follows is performed. First, the value PR.sub.n of the reference in-cylinder pressure is read from the storage section 42. Subsequently, the data (θ.sub.n, P.sub.n) is corrected so that the maximum value PR.sub.n.sub._.sub.max of the reference in-cylinder pressure and the maximum value P.sub.n.sub._.sub.max of the actual in-cylinder pressure correspond to each other (refer to the P.sub.max correction section 46a in FIG. 2). Subsequently, the data (θ.sub.n, P.sub.n) is further corrected so that the crank angle showing the maximum value P.sub.n.sub._.sub.max of the actual in-cylinder pressure corresponds to the TDC (refer to the TDC correction section 46b in FIG. 2).

[0082] Subsequently to step S12, the hysteresis zero angle H.sub.0 is determined (step S14). In the present step, processing as follows is performed. First, the deviation Δh.sub.n is calculated based on the correction data (θ.sub.n′, P.sub.n′) and the data of the reference in-cylinder pressure (θ.sub.n, PR.sub.n). In calculation of the deviation Δh.sub.n, equation (1) described above is used. Subsequently, the crank angle at a time of the value of the deviation Δh.sub.n changing from negative to positive is determined as the hysteresis zero angle H.sub.0 (the hysteresis zero angle determination section 46c in FIG. 2). Note that as explained with FIG. 6, the value of the deviation Δh.sub.n changes from negative to positive at the retardation side from the compression top dead center. Consequently, determination of the crank angle described above may be performed after the crank angle region is narrowed down to the crank angle region at the retardation side from the compression top dead center.

[0083] Subsequently to step S14, the gradient d.sub.n is calculated (step S16). In the present step, the gradient d.sub.n is calculated based on the data (θ.sub.n, Δh.sub.n) of the deviation Δh.sub.n at the retardation side from the hysteresis zero angle H.sub.0 and at the advance side from the predetermined crank angle (refer to the gradient calculation section 46d in FIG. 2). Any one of equations (4) to (6) described above is used in calculation of the gradient d.sub.n.

[0084] Subsequently to step S16, the correction coefficient η and the hysteresis zero angle H.sub.0 are updated (step S18). In the present step, the gradient d.sub.n calculated in step S16 is stored in the storage section 42 as the newest correction coefficient η. Further, the hysteresis zero angle H.sub.0 determined in step S14 is stored in the storage section 42 as the newest hysteresis zero angle H.sub.0.

[0085] When it is determined that the diesel engine is in the non-motoring state in step S10, the newest correction coefficient η and the hysteresis zero angle H.sub.0 are read from the storage section 42 (step S20).

[0086] Subsequently to step S20, the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure is corrected (step S22). In the present step, the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure, the newest correction coefficient η and hysteresis zero angle H.sub.0 are substituted into equation (7) as described above.

[0087] As above, according to the routine illustrated in FIG. 11, determination of the hysteresis zero angle H.sub.0 and calculation of the gradient d.sub.n can be performed from the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure in the motoring state. Further, the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure can be corrected by using the determined hysteresis zero angle H.sub.0 and the calculated gradient d.sub.n (that is, the correction coefficient η) in the non-motoring state. The hysteresis zero angle H.sub.0 and the correction coefficient η that are used in the non-motoring state are determined or calculated in a latest motoring state. Accordingly, precision of correction of the data of the actual in-cylinder pressure in the non-motoring state can be enhanced, and precision of combustion control of the diesel engine using the data can be enhanced.

[0088] Incidentally, in the above described first embodiment, the value of the deviation Δh.sub.n changes from negative to positive at the retardation side from the compression top dead center, and therefore, the crank angle at which the reversal occurs is determined as the hysteresis zero angle H.sub.0. However, since the cause of occurrence of difference in the gradient of the deviation Δh.sub.n is not certain as described in explanation of comparison of the lower tier in FIG. 7 and the lower tier in FIG. 8, there is a possibility that the crank angle at which the reversal of positive and negative of the value of the deviation Δh.sub.n occurs corresponds to the compression top dead center, or is at the advance side from the compression top dead center. However, it is needless to say that even in the case like this, if the crank angle at which the reversal of positive and negative of the value of the deviation Δh.sub.n occurs is determined as the hysteresis zero angle H.sub.0, the gradient d.sub.n can be calculated as in the above described first embodiment. Note that the present modification example can be similarly applied to a second embodiment that will be described later.

[0089] Further, in the above described first embodiment, the hysteresis zero angle H.sub.0 and the gradient d.sub.n are calculated by using the value of the deviation Δh.sub.n calculated in accordance with equation (1) described above. However, the hysteresis zero angle H.sub.0 and the gradient d.sub.n (that is, the rate of change of the pressure difference ΔP.sub.n) may be calculated by using the value of the pressure difference ΔP.sub.n that is calculated by equation (8) as follows that is obtained by simplifying equation (1) described above.

Pressure difference ΔP.sub.n=value P.sub.n of actual in-cylinder pressure-value PR.sub.n of reference in-cylinder pressure  (8)

In this connection, when the hysteresis zero angle H.sub.0 which is determined from equation (8) described above is used, the value P.sub.n of the actual in-cylinder pressure configuring the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure in the non-motoring state is corrected by equation (9) as follows.

Correction value of actual in-cylinder pressure=value P.sub.n of actual in-cylinder pressure×(detection crank angle θ.sub.n−hysteresis zero angle H.sub.0)×correction coefficient η   (9)

[0090] Note that the present modification example can be similarly applied to the second embodiment which will be described later.

Second Embodiment

[0091] Next, the second embodiment of the present application will be described with reference to FIGS. 12 to 14.

[0092] Note that a configuration of a diesel engine to which a control device according to the present embodiment is applied, and basic functions of the ECU 20 are common to the above described first embodiment, and therefore, explanation thereof will be omitted or simplified.

[0093] FIG. 12 is a functional block diagram of the ECU 20 as the control device according to the second embodiment of the present application. The basic functions of the ECU 20 are the same as the functions explained in FIG. 2. However, in FIG. 12, a gradient correction section 46e is provided. Further, in FIG. 12, the value Q of the fuel injection amount is outputted to not only the injector drive control section 38 but also the operation state determination section 32. In FIG. 12, the value Q of the fuel injection amount outputted to the operation state determination section 32 is also outputted to the storage section 42.

[0094] In the above described first embodiment, the gradient d.sub.n calculated in the motoring state is set as the correction coefficient η, and the correction coefficient η is used in correction of the data (θ.sub.n, P.sub.n) of the actual in-cylinder pressure in the non-motoring state. However, in the motoring state, the value Q of the fuel injection amount is zero, whereas in the cycle in the non-motoring state, the value Q of the fuel injection amount becomes larger than zero. Here, combustion is performed when the value Q of the fuel injection amount is larger than zero, and as the value Q of the fuel injection amount increases, a maximum value of the in-cylinder temperature at the time of combustion increases. When the maximum value of the in-cylinder temperature increases, viscosity of the deposits adhering to and accumulating on the periphery of the pressure receiving section of the in-cylinder pressure sensor 10 decreases. When the viscosity of the deposits decreases, the pressure receiving section 14 easily moves, and therefore the hysteresis that occurs to the value P.sub.n of the actual in-cylinder pressure reduces. That is, the difference between the “sensor value” and the “reference value” explained in FIG. 14 decreases. Therefore, the deviation Δh.sub.n that is calculated in equation (1) described above, and the gradient d.sub.n that is calculated in equations (4) to (6) described above decrease.

[0095] In the present embodiment, in order to take the characteristics by the fuel injection amount like this into consideration, correction that relates the gradient d.sub.n calculated in the motoring state to the fuel injection amount is performed in the gradient correction section 46e. FIG. 13 is a diagram explaining a method for relating the gradient d.sub.n to the fuel injection amount. The gradient d.sub.n which is calculated in the motoring state can be said as the correction coefficient η at the time of the value Q of the fuel injection amount being zero, and therefore, the value of the gradient d.sub.n is taken for an intercept of a vertical axis. When the fuel injection amount increases, the deviation Δh.sub.n becomes small, and therefore, both the value Q and the value of the gradient d.sub.n are related to each other so that as the value Q of the fuel injection amount becomes larger, the value of the gradient d.sub.0 becomes smaller (refer to equation (10) as follows).

Gradient d.sub.n=constant a×value Q of fuel injection amount+gradient d.sub.n.sub._.sub.Q=0  (10)

[0096] Further, the correction coefficient setting section 46 of the present embodiment outputs data (Q, d.sub.n) of the gradient to the storage section 42 as the correction coefficient η. When the storage section 42 receives data of the value Q of the fuel injection amount from the operation state determination section 32, the storage section 42 outputs the data of the correction coefficient η corresponding to the data to the in-cylinder pressure correction section 44. The in-cylinder pressure correction section 44 performs correction of the value P.sub.n of the actual in-cylinder pressure in the non-motoring state by using the data of the correction coefficient η and the hysteresis zero angle H.sub.0 which are received from the storage section 42. The value P.sub.n of the actual in-cylinder pressure in the non-motoring state is corrected by equation (7) described above.

[0097] According to the present embodiment described above, correction that relates the gradient d.sub.n calculated in the motoring state to the fuel injection amount is performed. Accordingly, precision of correction of the data of the actual in-cylinder pressure in the non-motoring state can be enhanced more.

[0098] Incidentally, in the above described second embodiment, the gradient d.sub.n calculated in the motoring state is related to the fuel injection amount. However, the gradient d.sub.n may be related to not only the fuel injection amount but also the engine speed in the motoring state. FIG. 14 is an example of a diagram illustrating the relationship illustrated in FIG. 13 at each engine speed in the motoring state. In the example illustrated in FIG. 14, an influence which the fuel injection amount gives to the gradient d.sub.n becomes smaller as the engine speed becomes higher.

[0099] In this case, the data (Q, Ne, d.sub.n) of the gradient is outputted to the storage section 42 as the correction coefficient η. Subsequently, when the storage section 42 receives the data of the value Q of the fuel injection amount and data of a value Ne of the engine speed, the storage section 42 outputs the data of the correction coefficient η corresponding to the data to the in-cylinder pressure correction section 44. The processing in the in-cylinder pressure correction section 44 which is performed hereinafter is the same as in the above described second embodiment.

[0100] From the above, when not only the fuel injection amount but also the engine speed in the motoring state influences the gradient d.sub.n, the engine speed is added to a basis of correction of the gradient d.sub.n, and thereby it becomes possible to enhance precision of correction of the data of the actual in-cylinder pressure in the non-motoring state more.