JP7509018B2 - Engine equipment - Google Patents

Description

The present invention relates to an engine device.

A conventional engine system of this type includes an engine and a fuel supply device having a fuel pump that supplies fuel in a fuel tank to a fuel passage connected to a fuel injection valve of the engine. When vapor generation in the fuel pump and fuel passage is detected, the rotation speed of the fuel pump is increased compared to when vapor generation is not detected (see, for example, Patent Document 1). In this engine system, this control causes vapor in the fuel pump and fuel passage to be discharged from the vapor discharge hole of the fuel pump.

JP 2015-48730 A

In the above-mentioned engine device, when the fuel in the fuel tank is boiling, depending on the fuel pump speed, vapor (vaporized fuel) generated in the fuel pump and fuel flow path may not be discharged sufficiently, or vapor may be generated in the fuel pump due to the amount of heat generated by the fuel pump.

The main objective of the engine device of the present invention is to be able to discharge vaporized fuel even if the fuel pump sucks in vaporized fuel when the fuel in the fuel tank is boiling, and to suppress the generation of vaporized fuel in the fuel pump due to the heat generated by the fuel pump.

The engine device of the present invention employs the following means to achieve the above-mentioned main objective.

The engine device of the present invention comprises:
The engine,
a fuel supply device having a fuel pump that supplies fuel in a fuel tank to a fuel passage connected to a fuel injection valve of the engine;
A control device for controlling the fuel pump;
An engine device comprising:
when the control device detects boiling of the fuel in the fuel tank, the control device sets a target rotation speed of the fuel pump within a range of a lower limit rotation speed at which the fuel pump can discharge the vaporized fuel when the vaporized fuel is sucked in by the fuel pump and an upper limit rotation speed at which generation of the vaporized fuel in the fuel pump due to an amount of heat generated by the fuel pump can be suppressed, and controls the fuel pump.
The gist of the present invention is as follows.

In the engine device of the present invention, when boiling of fuel in the fuel tank is detected, the target rotation speed of the fuel pump is set within a range between a lower limit rotation speed at which the fuel pump can discharge vaporized fuel (mainly fuel vaporized by boiling) when it draws in the vaporized fuel, and an upper limit rotation speed at which the generation of vaporized fuel in the fuel pump due to the heat generation of the fuel pump can be suppressed, and the fuel pump is controlled. As a result, even if the fuel pump draws in vaporized fuel when the fuel in the fuel tank is boiling, the vaporized fuel can be discharged, and the generation of vaporized fuel in the fuel pump due to the heat generation of the fuel pump can be suppressed.

In the engine device of the present invention, when the control device does not detect boiling of the fuel in the fuel tank, the control device sets the target rotation speed using feedback control of the proportional term and the integral term based on the difference between the fuel pressure, which is the pressure of the fuel in the fuel flow path, and the target fuel pressure, and further, the control device calculates the difference by subtracting a second processing value obtained by applying a smoothing process with a second response lower than the first response to the integral term from the integral term or a first processing value obtained by applying a smoothing process with a first response to the integral term, and detects boiling of the fuel in the fuel tank when the difference is equal to or greater than a difference threshold. When the fuel in the fuel tank boils and the fuel pump sucks in vaporized fuel and the discharge pressure of the fuel pump decreases, the fuel pressure becomes lower than the target fuel pressure, and the integral term in the feedback control increases. Because the integral term or the first processing value and the second processing value have different responsiveness (tracking ability) to the integral term, boiling of fuel in the fuel tank can be appropriately detected by using the difference condition obtained by subtracting the second processing value from the integral term or the first processing value.

In the engine device of the present invention, when the control device does not detect boiling of the fuel in the fuel tank, the control device sets the target rotation speed using feedback control using a proportional term and an integral term based on the difference between the fuel pressure, which is the pressure of the fuel in the fuel flow path, and the target fuel pressure, and further, the control device may detect boiling of the fuel in the fuel tank when a second difference obtained by subtracting the convergence integral term or the convergence processing value, which is the integral term or the first processing value when the fuel pressure convergence, which is the convergence of the fuel pressure, is detected from the integral term or the first processing value, is equal to or greater than a second difference threshold. When the fuel in the fuel tank is boiling and the fuel pump sucks in vaporized fuel and the discharge pressure of the fuel pump decreases, the fuel pressure becomes lower than the target fuel pressure, and the integral term in the feedback control increases. Therefore, by using the condition of the second difference obtained by subtracting the convergence integral term or the convergence processing value from the integral term or the first processing value, boiling of the fuel in the fuel tank can be appropriately detected.

In this case, the control device may detect the fuel pressure convergence when the fuel pressure difference between the target fuel pressure and the fuel pressure or the pulsation center fuel pressure, which is the pulsation center of the fuel pressure, is equal to or less than a fuel pressure difference threshold and continues for a predetermined time or more. When engine operation control is being performed, the fuel pressure pulsates due to the influence of fuel injection from the fuel injection valve, etc. For this reason, when the difference between the target fuel pressure and the pulsation center fuel pressure is used, it is easier to detect the convergence of the fuel pressure compared to when the difference between the target fuel pressure and the fuel pressure is used. In this case, the control device may estimate the pulsation center fuel pressure by performing an smoothing process on the fuel pressure.

The engine device of the present invention may further include a vaporized fuel processing device capable of purging the vaporized fuel gas containing vaporized fuel generated in the fuel tank to the intake pipe of the engine, and the control device may detect the boiling of fuel in the fuel tank when the purge concentration, which is the concentration of the purge, is higher than a concentration threshold value. In this way, the boiling of fuel in the fuel tank can be appropriately detected by using the purge concentration condition.

In the engine device of the present invention, when the control device does not detect boiling of the fuel in the fuel tank, the control device sets the target rotation speed using feedback control of the proportional term and the integral term based on the difference between the fuel pressure, which is the pressure of the fuel in the fuel flow path, and the target fuel pressure, and further, when the control device detects boiling of the fuel in the fuel tank, the control device may set the target rotation speed without using at least the integral term.

In this case, when the control device detects boiling of the fuel in the fuel tank, the control device may hold the integral term at the convergence time integral term, which is the integral term when fuel pressure convergence, which is the convergence of the fuel pressure, is detected. In this way, the integral term can be set to a more appropriate value when the control device subsequently resumes setting (updating) the integral term while assuming that boiling of the fuel in the fuel tank has not been detected.

In this case, when the boiling of fuel in the fuel tank is not detected (including when the control device is reset from detection to non-detection), when the target fuel pressure change amount, which is the absolute value of the change amount per unit time of the target fuel pressure, becomes equal to or exceeds the change amount threshold, the control device may hold the integral term until a release condition, including a condition that the release time has elapsed, is satisfied, and when the release condition is satisfied, the control device may resume setting (updating) the integral term. In this way, it is possible to prevent the time until the fuel pressure reaches the vicinity of the target fuel pressure from becoming longer. In this case, the control device may set the release time to be shorter when the target fuel pressure change amount becomes equal to or exceeds the change amount threshold due to a change from detection to non-detection of the boiling of fuel in the fuel tank compared to when the target fuel pressure change amount becomes equal to or exceeds the change amount threshold without being due to a change from detection to non-detection of the boiling of fuel in the fuel tank.

In the engine device of the present invention, the control device may stop the fuel pump from rotating when the engine is in a stopped state. This makes it possible to suppress noise and vibrations caused by driving the fuel pump.

1 is a schematic diagram showing the configuration of an automobile 10 equipped with an engine device 11 according to an embodiment of the present invention. 4 is a flowchart showing an example of a fuel injection control routine executed by an electronic control unit 70. 4 is a flowchart showing an example of a purge concentration related value setting routine executed by an electronic control unit 70. FIG. 4 is an explanatory diagram showing an example of an update amount setting map; 4 is a flowchart showing a part of an example of a fuel pump control routine executed by an electronic control unit 70. 4 is a flowchart showing a part of an example of a fuel pump control routine executed by an electronic control unit 70. 4 is a flowchart showing a part of an example of a fuel pump control routine executed by an electronic control unit 70. FIG. 13 is an explanatory diagram showing an example of a feedforward term setting map; FIG. 4 is an explanatory diagram showing an example of a target rotation speed setting map; 4 is a flowchart showing a part of an example of a fuel boiling detection flag setting routine executed by an electronic control unit 70. 4 is a flowchart showing a part of an example of a fuel boiling detection flag setting routine executed by an electronic control unit 70. 4 is a flowchart showing a part of an example of a fuel boiling detection flag setting routine executed by an electronic control unit 70. 4 is a flowchart showing an example of a fuel pressure convergence detection flag setting routine executed by an electronic control unit 70. 4 is an explanatory diagram showing an example of the target fuel pressure Pf*, the fuel pressure Pf, the feedback integral term αbi, the high response processing value βhi, the low response processing value βlo, the difference Δβ1, and the tentative detection flag Fv2b. 11 is an explanatory diagram showing an example of the target fuel pressure Pf*, the fuel pressure Pf, the feedback integral term αbi, the high response processing value βhi, the low response processing value βlo, the differences Δβ1, Δβ, and the tentative detection flags Fv2b, Fv2c. 4 is a flowchart showing a part of an example of a fuel pump control routine executed by an electronic control unit 70. 11 is an explanatory diagram showing an example of the target fuel pressure Pf*, the fuel pressure Pf, the feedback integral term αbi, the high response processing value βhi, the fuel pressure convergence detection flag Fpc, and the fuel boiling detection flag Fv. FIG. 13 is a flowchart illustrating an example of a feedback integral term setting process. FIG. 4 is an explanatory diagram showing an example of a first purge concentration threshold setting map; FIG. 4 is an explanatory diagram showing an example of an intake air temperature threshold setting map; FIG. 13 is an explanatory diagram showing an example of a second purge concentration threshold setting map; FIG. 11 is an explanatory diagram showing an example of a first difference threshold setting map; FIG. 13 is an explanatory diagram showing an example of a second difference threshold setting map;

Next, we will explain how to implement the present invention using examples.

Figure 1 is a schematic diagram showing the configuration of an automobile 10 equipped with an engine device 11 as one embodiment of the present invention. As shown in the figure, the automobile 10 of the embodiment is equipped with an engine 12, a fuel supply device 42, a vaporized fuel processing device 50, a transmission 60 connected to the crankshaft 14 of the engine 12 and connected to drive wheels 64a, 64b via a differential gear 62, a starter (not shown) for starting the engine 12, and an electronic control unit 70 for controlling the entire vehicle. The engine device 11 of the embodiment mainly includes the engine 12, the fuel supply device 42, the vaporized fuel processing device 50, and the electronic control unit 70.

The engine 12 is configured as an internal combustion engine that outputs power through four strokes: intake, compression, expansion, and exhaust, using fuel such as gasoline or diesel from a fuel tank 40. The engine 12 draws air cleaned by an air cleaner 22 into an intake pipe 23, passes it through a throttle valve 24, and injects fuel from a fuel injection valve 26 downstream of the throttle valve 24 of the intake pipe 23, mixing the air and fuel. The mixture is then drawn into a combustion chamber 29 via an intake valve 28, where it is explosively combusted by an electric spark from an ignition plug 30, and the reciprocating motion of a piston 32, which is pushed down by the energy of the explosive combustion, is converted into the rotational motion of a crankshaft 14. The exhaust gas discharged from the combustion chamber 29 into an exhaust pipe 34 via an exhaust valve 33 is discharged into the outside air via a purification device 35. The purification device 35 has a catalyst (three-way catalyst) 35a that purifies harmful components such as carbon monoxide (CO), hydrocarbons (HC), and nitrogen oxides (NOx).

The fuel supply device 42 is configured to be able to supply fuel in the fuel tank 40 to the fuel injection valve 26. The fuel supply device 42 includes a fuel flow path 43, a fuel pump 44, a relief passage 45, and a relief valve 46. The fuel flow path 43 is connected to the fuel injection valve 26. The fuel pump 44 supplies fuel in the fuel tank 40 to the fuel flow path 43. The fuel pump 44 is formed with a discharge hole 44a for discharging vaporized fuel (fuel vaporized by evaporation or boiling) generated in the fuel tank 40 by driving the fuel pump 44 when the vaporized fuel is sucked in.

The relief passage 45 is connected to the fuel flow passage 43 and the fuel tank 40. The relief valve 46 is provided in the relief passage 45, and closes when the fuel pressure (fuel pressure Pf) in the fuel flow passage 43 is less than the valve opening pressure Pfrv of the relief valve 46, and opens when the fuel pressure in the fuel flow passage 43 is equal to or greater than the valve opening pressure Pfrv. When the relief valve 46 opens, a portion of the fuel in the fuel flow passage 43 is returned to the fuel tank 40 via the relief passage 45. In this way, the fuel pressure in the fuel flow passage 43 is prevented from becoming excessive.

The vaporized fuel processing device 50 is configured to perform a purge that supplies vaporized fuel gas (purge gas) containing vaporized fuel generated in the fuel tank 40 to the intake pipe 23. The vaporized fuel processing device 50 includes an introduction passage 51, a shutoff valve 52, a canister 53, a purge passage 55, and a purge valve 56.

The introduction passage 51 is connected to the fuel tank 40 and the canister 53. The shutoff valve 52 is provided in the introduction passage 51 and is configured as a normally closed type electromagnetic valve. This shutoff valve 52 is controlled by the electronic control unit 70. The canister 53 is connected to the introduction passage 51 and the purge passage 55 and is open to the atmosphere via the atmosphere release passage 54. The inside of this canister 53 is filled with an adsorbent such as activated carbon that can adsorb vaporized fuel from the fuel tank 40. The atmosphere release passage 54 is provided with an air filter (not shown).

The purge passage 55 is connected to the canister 53 and the intake pipe 23 downstream of the throttle valve 24. The purge valve 56 is provided in the purge passage 55 and is configured as a normally closed type electromagnetic valve. This purge valve 56 is controlled by the electronic control unit 70.

This vaporized fuel processing device 50 performs purging by adjusting the opening of the purge valve 56 when the shutoff valve 52 is open. Specifically, vaporized fuel gas (purge gas) containing vaporized fuel generated in the fuel tank 40 is supplied to the intake pipe 23 through the introduction passage 51, the canister 53, and the purge passage 55 by utilizing the negative pressure in the intake pipe 23, with the flow rate being adjusted. Therefore, the engine 12 is capable of drawing a mixture of air and fuel (fuel from the fuel injection valve 26 and vaporized fuel from the vaporized fuel processing device 50) into the combustion chamber 29.

The electronic control unit 70 is configured as a microprocessor with a CPU at its core, and in addition to the CPU, it is equipped with a ROM for storing processing programs, a RAM for temporarily storing data, a flash memory for storing and holding data, and input/output ports. Signals from various sensors are input to the electronic control unit 70 via the input ports.

Examples of signals input to the electronic control unit 70 include the crank angle θcr from the crank position sensor 14a that detects the rotational position of the crankshaft 14 of the engine 12, and the cooling water temperature Tw from the water temperature sensor 15 that detects the temperature of the cooling water of the engine 12. Examples of signals input to the electronic control unit 70 include the throttle opening TH from the throttle position sensor 24a that detects the opening (position) of the throttle valve 24, and the cam angles θci and θco from the cam position sensor 16 that detects the rotational positions of the intake camshaft that opens and closes the intake valve 28 and the exhaust camshaft that opens and closes the exhaust valve 33. Examples of the air intake pressure include the intake air amount Qa from the air flow meter 23a attached upstream of the throttle valve 24 of the intake pipe 23, the intake air temperature Ta from the temperature sensor 23b attached upstream of the throttle valve 24 of the intake pipe 23, the front air-fuel ratio AF1 from the front air-fuel ratio sensor 37 attached upstream of the purifier 35 of the exhaust pipe 34, and the rear air-fuel ratio AF2 from the rear air-fuel ratio sensor 38 attached downstream of the purifier 35 of the exhaust pipe 34. Examples of the air-fuel ratio include the tank internal pressure Pt, which is the pressure inside the fuel tank 40, from the pressure sensor 40a attached to the fuel tank 40, the fuel pressure Pf, which is the pressure of the fuel inside the fuel flow passage 43, from the fuel pressure sensor 43p attached near the fuel injection valve 26 of the fuel flow passage 43 (for example, the delivery pipe), and the rotation speed Np of the fuel pump 44 from the rotation speed sensor 44b attached to the fuel pump 44. Examples of the sensor include the opening degree Osv of the shutoff valve 52 from the shutoff valve position sensor 52a that detects the opening degree (position) of the shutoff valve 52, and the opening degree Opv of the purge valve 56 from the purge valve position sensor 56a that detects the opening degree (position) of the purge valve 56. Examples of the sensor include an ignition signal IG from an ignition switch 80, and a shift position SP from a shift position sensor 82 that detects the operating position of a shift lever 81. Examples of the sensor include an accelerator opening Acc from an accelerator pedal position sensor 84 that detects the depression amount of an accelerator pedal 83, a brake pedal position BP from a brake pedal position sensor 86 that detects the depression amount of a brake pedal 85, a vehicle speed V from a vehicle speed sensor 88, and an atmospheric pressure Pout from an atmospheric pressure sensor 89.

Various control signals are output from the electronic control unit 70 via the output port. Examples of signals output from the electronic control unit 70 include a control signal to the throttle valve 24, a control signal to the fuel injector 26, a control signal to the spark plug 30, a control signal to the fuel pump 44, a control signal to the shutoff valve 52, and a control signal to the purge valve 56. Other examples include a control signal to the transmission 60 and a control signal to a starter (not shown).

The electronic control unit 70 calculates the rotation speed Ne of the engine 12 based on the crank angle θcr from the crank position sensor 14a. The electronic control unit 70 also calculates the load factor KL (the ratio of the volume of air actually taken in one cycle to the stroke volume per cycle of the engine 12) based on the intake air volume Qa from the air flow meter 23a and the rotation speed Ne of the engine 12. The electronic control unit 70 also calculates the consumption flow rate Qfr of the engine 12 based on the fuel injection amount Qf from the fuel injection valve 26.

In the automobile 10 of the embodiment thus configured, the electronic control unit 70 performs operation control of the engine 12 based on the required load rate KL* of the engine 12, which is based on the accelerator opening Acc and the vehicle speed V, specifically, intake air amount control that controls the opening of the throttle valve 24, fuel injection control that controls the amount of fuel injected from the fuel injection valve 26, and ignition control that controls the ignition timing of the spark plug 30. The electronic control unit 70 also controls the fuel pump 44 of the fuel supply device 42, and the shutoff valve 52 and purge valve 56 of the vaporized fuel processing device 50.

In the embodiment of the automobile 10, the electronic control unit 70 cuts fuel to the engine 12 when the accelerator is released while the automobile is traveling. Furthermore, in the embodiment of the automobile 10, when the automobile is stopped and an intermittent stop condition is met while the engine 12 is operating, the electronic control unit 70 intermittently stops the engine 12, and when a restart condition is met while the engine 12 is intermittently stopped, the starter cranks the engine 12 to restart the engine 12. For example, the intermittent stop condition is the accelerator being released and the brake being on. For example, the restart condition is the accelerator being on or the brake being released.

In the operation control of the engine 12, in the intake air amount control, the electronic control unit 70 sets the target opening TH* of the throttle valve 24 based on the required load rate KL* of the engine 12, and controls the throttle valve 24 using the set target opening TH*. In the ignition control, the electronic control unit 70 sets the target ignition timing Ti* of the spark plug 30 based on the rotation speed Ne and load rate KL of the engine 12, and controls the spark plug 30 using the set target ignition timing Ti*. In the control of the shutoff valve 52, when the tank internal pressure Pt (gauge pressure) is equal to or lower than a threshold value Ptref slightly higher than 0 kPa, the shutoff valve 52 is controlled to be in a closed state, and when the tank internal pressure Pt is higher than the threshold value Ptref, the shutoff valve 52 is controlled to be in an open state. In controlling the purge valve 56, when the purge conditions are met, the purge valve 56 is controlled to be in an open state and the above-mentioned purge is performed, and when the purge conditions are not met, the purge valve 56 is controlled to be in a closed state. As the purge conditions, for example, a condition is used in which the cooling water temperature Tw is equal to or higher than a threshold value, the operation control of the engine 12 (fuel injection control, etc.) is being performed, and the shutoff valve 52 is in an open state. The fuel injection control and the control of the fuel pump 44 are described below.

The fuel injection control will be explained. FIG. 2 is a flowchart showing an example of a fuel injection control routine executed by the electronic control unit 70. This routine is executed repeatedly when fuel injection control is performed. When this routine is executed, the electronic control unit 70 first inputs data such as the intake air amount Qa, the front air-fuel ratio AF1, the load rate KL, and the purge concentration related value Cpg (step S100). Here, the intake air amount Qa is input with a value detected by the air flow meter 23a. The front air-fuel ratio AF1 is input with a value detected by the front air-fuel ratio sensor 37. The load rate KL is input with a value calculated based on the intake air amount Qa and the rotation speed Ne as described above.

The purge concentration-related value Cpg is a value related to the deviation of the front air-fuel ratio AF1 from the required air-fuel ratio AF* (e.g., theoretical air-fuel ratio) per 1% of the purge rate (the ratio of vaporized fuel gas (purged gas) to the intake air volume). When the purge concentration-related value Cpg is a negative value, it means that the gas supplied to the combustion chamber 29 from the vaporized fuel processing device 50 through the intake pipe 23 contains vaporized fuel, and when the value is 0 or more, it means that the gas does not contain vaporized fuel. In addition, the purge concentration-related value Cpg becomes smaller (the absolute value becomes larger within the negative range) as the purge concentration (the concentration of vaporized fuel in the vaporized fuel gas when purging is performed) becomes higher. Furthermore, the purge concentration-related value Cpg is input with a value set by a purge concentration-related value setting routine executed by the electronic control unit 70, which will be described later.

When the data is input in this manner, the base injection amount Qfbs of the fuel injection valve 26 is set based on the load factor KL of the engine 12 (step S110). Here, the base injection amount Qfbs is the base value of the target injection amount Qf* of the fuel injection valve 26 for making the front air-fuel ratio AF1 the required air-fuel ratio AF*. This base injection amount Qfbs can be calculated, for example, as the value obtained by multiplying the unit injection amount Qfpu (fuel injection amount per 1% of the load factor KL) by the load factor KL.

Next, the purge correction coefficient Kpg is set based on the purge concentration related value Cpg (step S120), and the base injection amount Qfbs multiplied by the purge correction coefficient Kpg is set as the target injection amount Qf* (step S130). Then, the fuel injection valve 26 is controlled using the set target injection amount Qf* (step S140), and this routine ends. Here, the purge correction coefficient Kpg can be set, for example, by applying the purge concentration related value Cpg to a purge correction coefficient setting map. The purge correction coefficient setting map is predetermined as the relationship between the purge concentration related value Cpg and the purge correction coefficient Kpg, and is stored in the ROM or flash memory of the electronic control unit 70. The purge correction coefficient Kpg is set to be smaller relative to the value 1 as the purge concentration related value Cpg is smaller (the absolute value is larger within the negative range), that is, the purge concentration is higher. Therefore, the fuel injection valve 26 is controlled by setting the target injection amount Qf* to be smaller as the purge concentration is higher.

Next, the process of setting the purge concentration-related value Cpg used in the fuel injection control routine of FIG. 2 and the like will be described. FIG. 3 is a flowchart showing an example of a purge concentration-related value setting routine executed by the electronic control unit 70. This routine is executed repeatedly. Note that the purge concentration-related value Cpg and the number of purge concentration learning times Npg, which will be described later, are set to an initial value of 0, for example, at the start of a trip.

When the purge concentration related value setting routine in FIG. 3 is executed, the electronic control unit 70 first inputs data such as the front air-fuel ratio AF1 and the purge execution flag Fpg (step S200). Here, the value detected by the front air-fuel ratio sensor 37 is input as the front air-fuel ratio AF1.

The purge execution flag Fpg is set to a value of 1 when purging is being performed, and to a value of 0 when purging is not being performed. This purge execution flag Fpg is input with a value set by a purge execution flag setting routine (not shown) executed by the electronic control unit 70. In the purge execution flag setting routine, the electronic control unit 70 determines whether or not purging is being performed based on the opening degree Osv of the shutoff valve 52 and the opening degree Opv of the purge valve 56.

Once the data has been input in this manner, the value of the purge execution flag Fpg is checked (step S210). If the purge execution flag Fpg is equal to 0 (purge is not being performed), the purge concentration related value Cpg is retained (step S220), and the number of purge concentration learning attempts Npg is retained (step S230), after which the routine ends.

When the purge execution flag Fpg is set to a value of 1 in step S210 (purging is being performed), the update amount ΔCpg of the purge concentration related value Cpg is set based on the front air-fuel ratio AF1 (step S240), and the purge concentration related value Cpg is updated (learned) by adding the update amount ΔCpg to the previous value of the purge concentration related value Cpg and setting the new purge concentration related value Cpg to the value (step S250), and the purge concentration learning count Npg is counted up by a value of 1 (step S260), and this routine ends.

Here, the update amount ΔCpg can be set, for example, by applying the front air-fuel ratio AF1 to the update amount setting map. The update amount setting map is determined in advance as the relationship between the front air-fuel ratio AF1 and the update amount ΔCpg, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 4 is an explanatory diagram showing an example of the update amount setting map. As shown in the figure, the update amount ΔCpg is set to a value of 0 when the front air-fuel ratio AF1 is equal to the required air-fuel ratio AF*. In addition, when the front air-fuel ratio AF1 is larger than the required air-fuel ratio AF* (when on the lean side), the update amount ΔCpg is set to be larger within the positive range as the front air-fuel ratio AF1 is larger. Furthermore, when the front air-fuel ratio AF1 is smaller than the required air-fuel ratio AF* (when on the rich side), the update amount ΔCpg is set to be smaller within the negative range (the absolute value becomes larger) as the front air-fuel ratio AF1 is smaller. Basically, the higher the purge concentration, the more fuel there is in the combustion chamber 29, and the more likely the front air-fuel ratio AF1 is to be smaller than the required air-fuel ratio AF*. Therefore, the higher the purge concentration, the more likely the update amount ΔCpg and therefore the purge concentration-related value Cpg are to be smaller (the absolute value tends to be larger within the negative range).

Next, the control of the fuel pump 44 of the fuel supply device 42 will be described. Figures 5 to 7 are flowcharts showing an example of a fuel pump control routine executed by the electronic control unit 70. This routine is executed repeatedly. When this routine is executed, the electronic control unit 70 first inputs data such as the fuel pressure Pf, the load factor KL, the engine rotation flag Fer, the fuel cut flag Ffc1, and the fuel boiling detection flag Fv (step S300). Here, the value detected by the fuel pressure sensor 43p is input as the fuel pressure Pf. The value calculated based on the intake air amount Qa and the rotation speed Ne as described above is input as the load factor KL.

The engine rotation flag Fer is set to a value of 1 when the engine 12 is rotating and to a value of 0 when the engine 12 is stopped. This engine rotation flag Fer receives a value set by an engine rotation flag setting routine (not shown) executed by the electronic control unit 70. In the engine rotation flag setting routine, the electronic control unit 70 determines whether the engine 12 is rotating or stopped based on the engine 12 rotation speed Ne.

The fuel cut flag Ffc1 is set to a value of 1 when the engine 12 is being fuel cut, and is set to a value of 0 when the engine 12 is not being fuel cut (fuel injection control, etc. is being performed). This fuel cut flag Ffc1 is input with a value set by a fuel cut flag setting routine (not shown) executed by the electronic control unit 70.

The fuel boiling detection flag Fv is set to a value of 1 when boiling of fuel in the fuel tank 40 is detected, and is set to a value of 0 when boiling of fuel in the fuel tank 40 is not detected (including when it is reset from detection to not detection). This fuel boiling detection flag Fv is input with a value set by a fuel boiling detection flag setting routine executed by the electronic control unit 70, which will be described later. Note that the fuel in the fuel tank 40 boils when the absolute pressure in the fuel tank 40 is equal to or lower than the saturated vapor pressure of the fuel. Therefore, the fuel in the fuel tank 40 is likely to boil when the altitude is high and the atmospheric pressure Pout is low, or when the temperature of the fuel in the fuel tank 40 is high and the saturated vapor pressure of the fuel is high.

Once the data has been input in this manner, the value of the engine rotation flag Fer is checked (step S310). When the engine rotation flag Fer is equal to 1 (the engine 12 is rotating), the value of the fuel boiling detection flag Fv is checked (step S320). When the fuel boiling detection flag Fv is equal to 0 (the boiling of fuel in the fuel tank 40 has not been detected), the target fuel pressure Pf* is set within a range less than the opening pressure Pfrv of the relief valve 46 based on the state of the engine 12, etc. (step S330). The target fuel pressure Pf* may be used for controlling the operation of the engine 12, in addition to controlling the fuel pump 44.

Then, the value of the fuel cut flag Ffc1 is checked (step S340). When the fuel cut flag Ffc1 is set to 0 (fuel cut of the engine 12 is not being performed), the feedforward term αf used to set the target discharge flow rate Qp* of the fuel pump 44 and the feedback proportional term αbp and feedback integral term αbi, which are the proportional and integral terms of the feedback term, are set by fuel pressure feedback control (steps S350 to S354).

First, the feedforward term αf is set based on the load rate KL (step S350). Here, the feedforward term αf can be set, for example, by applying the load rate KL to a feedforward term setting map. The feedforward term setting map is determined in advance as a relationship between the load rate KL and the feedforward term αf, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 8 is an explanatory diagram showing an example of a feedforward term setting map. As shown in the figure, the feedforward term αf is set so that it increases as the load rate KL increases. Next, as shown in equation (1), the product of the value obtained by subtracting the fuel pressure Pf from the target fuel pressure Pf* and the gain Kp of the proportional term is set to the feedback proportional term αbp (step S352). Then, as shown in equation (2), the feedback integral term αbi is updated by setting the new feedback integral term αbi to the sum of the previous value of the feedback integral term αbi and the product of the value obtained by subtracting the fuel pressure Pf from the target fuel pressure Pf* and the gain Ki of the integral term (step S354).

αbp＝Kp・(Pf*－Pf) (1)
αbi = previous αbi + Ki · (Pf* - Pf) (2)

When the fuel cut flag Ffc1 is set to a value of 1 in step S340 (fuel cut is being performed on the engine 12), the feedforward term αf is set to a value of 0 (step S360), the feedback proportional term αbp is set using equation (1) as in the processing of step S352 (step S362), and the feedback integral term αbi is held at its previous value (step S364).

After setting the feedforward term αf, the feedback proportional term αbp, and the feedback integral term αbi in this manner, the sum of these is lower-limit guarded at 0 and set as the target discharge flow rate Qp* of the fuel pump 44 (step S370). Next, the target rotation speed Np* of the fuel pump 44 is set based on the set target discharge flow rate Qp* (step S372), the fuel pump 44 is controlled using the set target rotation speed Np* (step S374), and this routine ends.

In this case, the target rotation speed Np* can be set, for example, by applying the target discharge flow rate Qp* to a target rotation speed setting map. The target rotation speed setting map is determined in advance as the relationship between the target discharge flow rate Qp* and the target rotation speed Np*, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 9 is an explanatory diagram showing an example of a target rotation speed setting map. As shown in the figure, the target rotation speed Np* is set to be higher as the target discharge flow rate Qp* increases. Hereinafter, the control of the fuel pump 44 (steps S330 to S374) when the engine rotation flag Fer is set to a value of 1 and the fuel boiling detection flag Fv is set to a value of 0 will be referred to as "normal pump control."

When the fuel boiling detection flag Fv is set to a value of 1 in step S320 (fuel boiling in the fuel tank 40 is detected), the relief valve 46 opening pressure Pfrv is set to the target fuel pressure Pf* (step S380), and the value of the fuel cut flag Ffc1 is checked (step S390). When the fuel cut flag Ffc1 is set to a value of 0 (fuel cut of the engine 12 is not being performed), the feedforward term αf is set (step S392) in the same manner as in the processing of step S350 described above. When the fuel cut flag Ffc1 is set to a value of 1 (fuel cut of the engine 12 is being performed), the feedforward term αf is set to a value of 0 (step S394).

When the feedforward term αf is set in this manner, the set feedforward term αf is set to the target discharge flow rate Qp* of the fuel pump 44 (step S400), and similarly to the processing of step S372 described above, a provisional rotation speed Nptmp, which is a provisional value of the target rotation speed Np* of the fuel pump 44, is set based on the set target discharge flow rate Qp* (step S402). In this case, the provisional rotation speed Nptmp can be set, for example, by applying the target discharge flow rate Qp* to a map in which the vertical axis of the target rotation speed setting map in FIG. 9 is replaced from "target rotation speed Np*" to "provisional rotation speed Nptmp".

Next, as shown in equation (3), the set tentative rotation speed Nptmp of the fuel pump 44 is set to a target rotation speed Np* of the fuel pump 44, which is a value obtained by limiting the set provisional rotation speed Nptmp with the lower limit rotation speed Npmin and the upper limit rotation speed Npmax (step S404). The set target rotation speed Np* is then used to control the fuel pump 44 (step S406). Details of the lower limit rotation speed Npmin and the upper limit rotation speed Npmax will be described later. Hereinafter, the control of the fuel pump 44 when the engine rotation flag Fer is set to a value of 1 and the fuel boiling detection flag Fv is set to a value of 1 (steps S380 to S406) is referred to as "pump control during boiling".

Np* = min(max(Nptmp, Npmin), Npmax) (3)

When the fuel in the fuel tank 40 is boiling, the fuel pump 44 may suck in vaporized fuel (fuel vaporized by evaporation or boiling, mainly the latter), causing the discharge pressure of the fuel pump 44 to drop, or the discharge pressure of the fuel pump 44 has already dropped. If the discharge pressure of the fuel pump 44 drops, it will lead to a drop in the fuel pressure Pf and therefore a drop in the amount of fuel injected from the fuel injection valve 26 (a drop in the controllability of the fuel injection control). For this reason, it is necessary to suppress the drop in the discharge pressure of the fuel pump 44, or to eliminate it if it has already dropped.

Based on this, in the embodiment, when the fuel boiling detection flag Fv is set to a value of 1 (boiling of fuel in the fuel tank 40 is detected), pump control during boiling is executed. Specifically, the fuel pump 44 is controlled by setting the provisional rotation speed Nptmp of the fuel pump 44 to a target rotation speed Np*, which is a value that is upper and lower limit guarded by the lower limit rotation speed Npmin and the upper limit rotation speed Npmax. As the lower limit rotation speed Npmin, the lower limit value of the rotation speed range of the fuel pump 44 is used, which can sufficiently discharge the vaporized fuel (mainly fuel vaporized by boiling) from the discharge hole 44a when the fuel pump 44 sucks the vaporized fuel. As the upper limit rotation speed Npmax, the upper limit value of the rotation speed range of the fuel pump 44 is used, which can sufficiently suppress the generation of vaporized fuel in the fuel pump 44 due to the heat generation of the fuel pump 44.

When the fuel in the fuel tank 40 is boiling, the load factor KL is low or fuel is being cut off, the feedforward term αf is reduced and the provisional rotation speed Nptmp of the fuel pump 44 is reduced (steps S390 to S402). In this embodiment, the provisional rotation speed Nptmp of the fuel pump 44 is set to a value that is a lower limit guard value set by the lower limit rotation speed Npmin as the target rotation speed Np*, so that when the fuel pump 44 sucks in vaporized fuel (mainly fuel vaporized by boiling), the vaporized fuel can be sufficiently discharged from the discharge hole 44a. As a result, a decrease in the discharge pressure of the fuel pump 44 can be suppressed, or a decrease can be eliminated if it has already occurred.

In addition, when the fuel in the fuel tank 40 is boiling, if the rotation speed Np of the fuel pump 44 is set too high, the amount of heat generated by the fuel pump 44 may cause vaporized fuel to be generated in the fuel pump 44. In the embodiment, the temporary rotation speed Nptmp of the fuel pump 44 is set to a value that is an upper limit guard value of the upper limit rotation speed Npmax, and the generation of vaporized fuel in the fuel pump 44 due to the amount of heat generated by the fuel pump 44 can be sufficiently suppressed.

When boiling pump control is executed in this manner (steps S380 to S406), the integrated execution time Tfup of boiling pump control is measured (step S408), the feedback integral term αbi is held at the previous value (step S410), and this routine ends. Here, the integrated execution time Tfup is set to an initial value of 0, for example, when the automobile 10 is shipped or during maintenance. In the embodiment, the integrated execution time Tfup is stored in the RAM or flash memory of the electronic control unit 70, or transmitted to the manufacturer, dealer, etc. This allows the manufacturer, dealer, etc. to use the integrated execution time Tfup to perform various analyses, etc.

When the engine rotation flag Fer is 0 in step S310 (the engine 12 is in a rotation-stopped state), the target fuel pressure Pf*, the target discharge flow rate Qp*, and the target rotation speed Np* are each set to 0 (steps S420 to S424), and the fuel pump 44 is controlled using the set target rotation speed Np* (step S426). Then, the feedback integral term αbi is held at the previous value (step S428), and this routine is terminated. In this case, the fuel pump 44 is stopped (changing from a rotating state to a rotation-stopped state or maintaining the rotation-stopped state). In this way, when the engine 12 is in a rotation-stopped state, the fuel pump 44 is stopped regardless of the value of the fuel boiling detection flag Fv (regardless of whether the boiling of the fuel in the fuel tank 40 has been detected or not). This makes it possible to suppress the noise and vibration caused by the operation of the fuel pump 44 from being felt by the occupants (including the driver).

Next, the process for setting the fuel boiling detection flag Fv used in the fuel pump control routine of FIG. 5 to FIG. 7 and the process for setting the fuel pressure convergence detection flag Fpc used in the process for setting the fuel boiling detection flag Fv will be described. Here, the fuel pressure convergence detection flag Fpc is set to a value of 1 when the fuel pressure Pfcn, which is the center of the pulsation of the fuel pressure Pf, is detected as converging near the target fuel pressure Pf*, and is set to a value of 0 when the fuel pressure convergence is not detected (including when the fuel pressure convergence is reset from detection to not detection). The value set by the fuel pressure convergence detection flag setting routine executed by the electronic control unit 70 is input to this fuel pressure convergence detection flag Fpc. The fuel boiling detection flag Fv and the fuel pressure convergence detection flag Fpc are set to an initial value of 0, for example, at the start of a trip.

FIGS. 10 to 12 are flowcharts showing an example of a fuel boiling detection flag setting routine executed by the electronic control unit 70, and FIG. 13 is a flowchart showing an example of a fuel pressure convergence detection flag setting routine executed by the electronic control unit 70. Each of these routines is executed repeatedly. For ease of explanation, the fuel pressure convergence detection flag setting routine in FIG. 13 will be explained first, followed by the fuel boiling detection flag setting routine in FIG. 10 to 12.

The fuel pressure convergence detection flag setting routine in Figure 13 will now be described. When this routine is executed, the electronic control unit 70 first inputs data such as the fuel pressure Pf, the target fuel pressure Pf*, the fuel pump rotation flag Fpr, and the fuel cut continuation flag Ffc2 (step S700). Here, the value detected by the fuel pressure sensor 43p is input as the fuel pressure Pf. The value set by the fuel pump control routine in Figures 5 to 7 is input as the target fuel pressure Pf*.

The fuel pump rotation flag Fpr is a flag that is set to a value of 1 when the fuel pump 44 is rotating and to a value of 0 when the fuel pump 44 is stopped. This fuel pump rotation flag Fpr receives a value set by a fuel pump rotation flag setting routine (not shown) executed by the electronic control unit 70. In the fuel pump rotation flag setting routine, the electronic control unit 70 determines whether the fuel pump 44 is rotating or stopped based on the rotation speed Np of the fuel pump 44.

The fuel cut continuation flag Ffc2 is set to a value of 1 when the condition that the fuel cut of the engine 12 is being performed and the duration of the cut is equal to or longer than a predetermined time Ffc is met, and is set to a value of 0 when this condition is not met. This fuel cut continuation flag Ffc2 is input with a value set by a fuel cut continuation flag setting routine (not shown) executed by the electronic control unit 70. As the predetermined time Ffc, a time at which the reliability of the feedback integral term αbi can no longer be guaranteed, for example, several hundred to several thousand msec, is used.

When the data is input in this manner, the value of the fuel pump rotation flag Fpr is checked (step S710), the target fuel pressure Pf* is compared with the opening pressure Pfrv of the relief valve 46 (step S712), and the value of the fuel cut continuation flag Ffc2 is checked (step S714). As explained in the fuel pump control routine of Figures 5 to 7, when the fuel pump 44 is in a rotation stop state, when the target fuel pressure Pf* is equal to the opening pressure Pfrv, or when fuel cut is being performed, the feedback integral term αbi is held. This reduces the reliability of the feedback integral term αbi. The processing of steps S710 to S714 is a processing for determining whether or not the fuel pressure convergence is undetected (holding as undetected or resetting from detection to undetected) based on these facts.

When the fuel pump rotation flag Fpr is equal to 0 in step S710 (the fuel pump 44 is stopped), when the target fuel pressure Pf* is equal to the relief valve 46 opening pressure Pfrv in step S712, or when the fuel cut continuation flag Ffc2 is equal to 1 in step S714 (fuel cut is being performed and its duration is equal to or longer than the predetermined time Ffc), it is determined that fuel pressure convergence has not been detected, the fuel pressure convergence detection flag Fpc is set to 0 (step S716), and this routine is terminated.

When the fuel pump rotation flag Fpr is set to 1 in step S710 (the fuel pump 44 is rotating), and the target fuel pressure Pf* is less than the relief valve opening pressure Pfrv in step S712, and the fuel cut continuation flag Ffc2 is set to 0 in step S714 (fuel cut is not being performed or the duration of fuel cut is less than the predetermined time Ffc), the fuel pressure Pf is smoothed and the value obtained is estimated as the pulsation center fuel pressure Pfcn (step S720) without determining that fuel pressure convergence has not been detected. When the engine 12 is being controlled, the fuel pressure Pf pulsates due to the influence of fuel injection from the fuel injection valve 26, etc. The process of step S720 is a process of estimating the pulsation center of this fuel pressure Pf.

Next, the absolute value of the value obtained by subtracting the pulsation center fuel pressure Pfcn from the target fuel pressure Pf* is set as the fuel pressure difference ΔPf (step S730), and it is determined whether or not the condition that the fuel pressure difference ΔPf is equal to or less than the threshold value ΔPfref and the duration of the difference is equal to or more than the predetermined time Tpf is satisfied (steps S740, S742). Here, the threshold value ΔPfref is used to determine whether the fuel pressure difference ΔPf is sufficiently small. The predetermined time Tpf is the time required to determine that the fuel pressure difference ΔPf is stable at or less than the threshold value ΔPfref (the pulsation center fuel pressure Pfcn is converging near the target fuel pressure Pf*), for example, several hundred msec. The processing of steps S740 and S742 is a processing to determine whether the fuel pressure convergence is detected (change from undetected to detected or maintain detected).

In steps S740 and S742, when the condition that the fuel pressure difference ΔPf is equal to or less than the threshold value ΔPfref and the duration is equal to or more than the predetermined time Tpf is satisfied, it is determined that the fuel pressure convergence is detected, the fuel pressure convergence detection flag Fpc is set to a value of 1 (step S744), and this routine is terminated. As described above, when the engine 12 is being controlled, the fuel pressure Pf pulsates due to the influence of fuel injection from the fuel injection valve 26. Therefore, by setting the fuel pressure difference ΔPf to the absolute value of the value obtained by subtracting the pulsation center fuel pressure Pfcn from the target fuel pressure Pf*, it is possible to more easily detect the fuel pressure convergence than by setting the fuel pressure difference ΔPf to the absolute value of the value obtained by subtracting the fuel pressure Pf from the target fuel pressure Pf*. In addition, when the fuel pressure Pf and the pulsation center fuel pressure Pfcn are converging near the target fuel pressure Pf*, the feedback integral term αbi, which is sequentially updated in the fuel pump control routine of FIG. 5 to FIG. 7 when the engine rotation flag Fer is set to a value of 1, the fuel boiling detection flag Fv is set to a value of 0, and the fuel cut flag Ffc1 is set to a value of 1, is also assumed to be stable.

If the condition that the fuel pressure difference ΔPf is equal to or less than the threshold value ΔPfref and the duration of this difference is equal to or greater than the predetermined time Tpf is not met in steps S740 and S742, the fuel pressure convergence detection flag Fpc is held at the previous value without determining that the fuel pressure is to be detected (step S746), and the routine ends.

In the embodiment, when the fuel pump rotation flag Fpr is 0 (the fuel pump 44 is in a stopped state), when the target fuel pressure Pf* is equal to the opening pressure Pfrv of the relief valve 46, or when the fuel cut continuation flag Ffc2 is 1 (fuel cut is being performed and its duration is equal to or longer than the predetermined time Ffc), the fuel pressure convergence detection flag Fpc is set to 0 (fuel pressure convergence is not detected). Therefore, when the fuel pressure convergence detection flag Fpc is 1, the fuel boiling detection flag Fv is switched to switch the target fuel pressure Pf*, or when the duration of fuel cut is less than the predetermined time Tfc, the fuel pressure convergence detection flag Fpc is held at 1.

Next, the fuel boiling detection flag setting routine of Figures 10 to 12 will be described. When this routine is executed, the electronic control unit 70 first inputs data such as the cooling water temperature Tw, intake air temperature Ta, tank internal pressure Pt, purge concentration related value Cpg, purge concentration learning count Npg, feedback integral term αbi, integral term update continuation flag Fiu, consumption flow rate stable flag Ffr, and fuel pressure convergence detection flag Fpc (step S500).

Here, the value detected by the water temperature sensor 15 is input as the cooling water temperature Tw. The value detected by the temperature sensor 23b is input as the intake air temperature Ta. The value detected by the pressure sensor 40a is input as the tank internal pressure Pt. The values set by the purge concentration related value setting routine in FIG. 3 are input as the purge concentration related value Cpg and the number of purge concentration learnings Npg. The value set by the fuel pump control routine in FIG. 5 to FIG. 7 is input as the feedback integral term αbi.

The integral term update continuation flag Fiu is set to a value of 1 when the feedback integral term αbi is updated and the update duration is equal to or longer than a predetermined time Tiu, and is set to a value of 0 when this condition is not met. The integral term update continuation flag Fiu is input with a value set by an integral term update continuation flag setting routine (not shown) executed by the electronic control unit 70. As the predetermined time Tiu, a time that can ensure the reliability of the feedback integral term αbi, for example, several hundred to several thousand msec, is used.

The consumption flow rate stability flag Ffr is set to a value of 1 when the fuel consumption flow rate Qfr of the engine 12 is stable, and is set to a value of 0 when the fuel consumption flow rate Qfr of the engine 12 is not stable. The consumption flow rate stability flag Ffr is input with a value set by a consumption flow rate stability flag setting routine (not shown) executed by the electronic control unit 70. In the consumption flow rate stability flag setting routine, the electronic control unit 70 determines that the fuel consumption flow rate Qfr of the engine 12 is stable when the consumption flow rate fluctuation amount ΔQfr (the difference between the maximum value and the minimum value), which is the fluctuation amount per unit time of the consumption flow rate Qfr, is equal to or less than the threshold value ΔQfrref and the duration of the fluctuation amount is equal to or more than the predetermined time Tfr. When this condition is not satisfied, the electronic control unit 70 determines that the fuel consumption flow rate Qfr of the engine 12 is not stable. The consumption flow rate Qfr is calculated based on the fuel injection amount Qf from the fuel injection valve 26. The threshold value ΔQfcref is used to determine whether the consumption flow rate fluctuation amount ΔQfr is sufficiently small. The predetermined time Tfr is the time required to determine that the consumption flow rate fluctuation amount ΔQfr is stable at or below the threshold value ΔQfrref, for example, several hundred to several thousand msec.

When the data is input in this manner, the value obtained by performing high-response smoothing processing on the feedback integral term αbi is set to the high-response processing value βhi (step S510), and the value obtained by performing low-response smoothing processing on the feedback integral term αbi is set to the low-response processing value βlo (step S512). Here, the high-response smoothing processing is smoothing processing using a predetermined number of times Nsm1 (e.g., several times) as the smoothing number Nsm, and the low-response smoothing processing is smoothing processing using a predetermined number of times Nsm2 (e.g., several thousand times) that is greater than the predetermined number of times Nsm1 as the smoothing number Nsm.

Next, it is determined whether or not the fuel pressure convergence detection flag Fpc has just switched from 0 to 1 (step S514). If the fuel pressure convergence detection flag Fpc has just switched from 0 to 1 (immediately after fuel pressure convergence has changed from undetected to detected), the high response processing value βhi is set to the convergence processing value βhi0 (step S516). If the fuel pressure convergence detection flag Fpc has not just switched from 0 to 1 (immediately after fuel pressure convergence has changed from undetected to detected), the process of step S516 is not executed. Note that in this embodiment, the convergence processing value βhi0 is reset when the fuel pressure convergence detection flag Fpc switches from 1 to 0.

Then, the purge concentration learning count Npg is compared with the threshold value Npgref (step S520), the purge concentration related value Cpg is compared with the threshold value Cpgref1 (step S522), and the tank internal pressure Pt is compared with the threshold value Ptref2 (step S524). Here, the threshold value Npgref is used to determine whether the reliability of the purge concentration related value Cpg can be guaranteed. The threshold value Cpgref1 is used to determine whether the purge concentration related value Cpg is relatively large, i.e., whether the purge concentration is relatively low. The threshold value Ptref is used to determine whether the tank internal pressure Pt is about 0 kPa. When the fuel in the fuel tank 40 is boiling, more vaporized fuel is generated in the fuel tank 40 than when the fuel is not boiling, and the tank internal pressure Pt is likely to be high. In addition, it is assumed that the smaller the purge concentration related value Cpg (the larger the absolute value within the negative range), i.e., the higher the purge concentration, the more vaporized fuel is generated in the fuel tank 40. The processing of steps S520 to S524 is a process that, based on these, uses the purge concentration learning count Npg, the purge concentration related value Cpg, and the tank internal pressure Pt to determine whether or not the boiling of fuel in the fuel tank 40 is undetected (remaining undetected or resetting from detected to undetected).

When the purge concentration learning count Npg is equal to or greater than the threshold value Npgref in step S520, the purge concentration related value Cpg is equal to or greater than the threshold value Cpgref1 in step S522, and the tank internal pressure Pt is less than the threshold value Ptref in step S524, it is determined that the boiling of fuel in the fuel tank 40 is undetected based on the conditions of the purge concentration learning count Npg, the purge concentration related value Cpg, and the tank internal pressure Pt, and the provisional undetected flag Fv1a is set to a value of 1 (step S526).

When the purge concentration learning count Npg is less than the threshold value Npgref in step S520, when the purge concentration related value Cpg is less than the threshold value Cpgref1 in step S522, or when the tank internal pressure Pt is equal to or greater than the threshold value Ptref in step S524, the provisional undetected flag Fv1a is set to a value of 0 without determining that the boiling of fuel in the fuel tank 40 is undetected under the conditions of the purge concentration learning count Npg, the purge concentration related value Cpg, and the tank internal pressure Pt (step S528).

The cooling water temperature Tw is then compared with a threshold value Twref (step S530). Here, the threshold value Twref is used to determine whether the cooling water temperature Tw is relatively low or not. In general, when the cooling water temperature Tw is low, the temperature of the fuel in the fuel tank 40 is also low, and it is assumed that the fuel in the fuel tank 40 is unlikely to boil. Based on this, the process of step S530 is a process that uses the cooling water temperature Tw to determine whether the boiling of the fuel in the fuel tank 40 is undetected or not.

When the cooling water temperature Tw is less than the threshold value Twref in step S530, it is determined that the boiling of fuel in the fuel tank 40 is undetected under the cooling water temperature Tw condition, and the provisional undetection flag Fv1b is set to a value of 1 (step S532). When the cooling water temperature Tw is equal to or greater than the threshold value Twref, it is not determined that the boiling of fuel in the fuel tank 40 is undetected under the cooling water temperature Tw condition, and the provisional undetection flag Fv1b is set to a value of 0 (step S534).

The intake temperature Ta is also compared with a threshold value Taref (step S540). Here, the threshold value Taref is used to determine whether the intake temperature Ta is relatively low or not. In general, when the intake temperature Ta is low, the temperature of the fuel in the fuel tank 40 is also low, and it is assumed that the fuel in the fuel tank 40 is unlikely to boil. Based on this, the process of step S540 is a process that uses the intake temperature Ta to determine whether the boiling of the fuel in the fuel tank 40 is undetected or not.

When the intake temperature Ta is less than the threshold value Taref in step S540, it is determined that the boiling of fuel in the fuel tank 40 is undetected under the intake temperature Ta condition, and the provisional undetection flag Fv1c is set to a value of 1 (step S542). When the intake temperature Ta is equal to or greater than the threshold value Taref, it is not determined that the boiling of fuel in the fuel tank 40 is undetected under the intake temperature Ta condition, and the provisional undetection flag Fv1c is set to a value of 0 (step S544).

When the provisional undetected flags Fv1a, Fv1b, and Fv1c are set in this way, the values of the provisional undetected flags Fv1a, Fv1b, and Fv1c that have been set are checked (steps S550 to S554). When at least one of the provisional undetected flags Fv1a, Fv1b, and Fv1c is set to a value of 1, the fuel boiling detection flag Fv is set to a value of 0 (step S556), and this routine is terminated. That is, when it is determined that the boiling of fuel in the fuel tank 40 is undetected based on at least one of the conditions of the purge concentration learning count Npg, the purge concentration related value Cpg, and the tank internal pressure Pt (steps S520 to S524), the condition of the cooling water temperature Tw (step S530), and the condition of the intake air temperature Ta (step S540), the value of 0 is set to the fuel boiling detection flag Fv1. In this way, the boiling of fuel in the fuel tank 40 can be determined to be undetected (retained as undetected or reset from detection to undetected).

When all of the provisional undetected flags Fv1a, Fv1b, and Fv1c are set to 0 in steps S550 to S554, the process proceeds to step S560. That is, when it is not determined that the boiling of fuel in the fuel tank 40 is undetected under any of the conditions of the purge concentration learning count Npg, the purge concentration related value Cpg, and the tank internal pressure Pt (steps S520 to S524), the condition of the cooling water temperature Tw (step S530), and the condition of the intake air temperature Ta (step S540), the process proceeds to step S560.

Next, the purge concentration-related value Cpg is compared with a threshold value Cpgref2 that is equal to or less than the threshold value Cpgref1 described above (step S560). Here, the threshold value Cpgref2 is used to determine whether the purge concentration-related value Cpg is relatively small (the absolute value is large within the negative range), i.e., whether the purge concentration is relatively high. As described above, when the fuel in the fuel tank 40 is boiling, more vaporized fuel is generated in the fuel tank 40 than when the fuel is not boiling. In addition, it is assumed that the smaller the purge concentration-related value Cpg is (the larger the absolute value is within the negative range), i.e., the higher the purge concentration is, the more vaporized fuel is generated in the fuel tank 40. Based on this, the process of step S560 is a process of determining whether or not boiling of the fuel in the fuel tank 40 is detected using the purge concentration learning count Npg.

When the purge concentration-related value Cpg is less than the threshold value Cpgref2 in step S560, it is determined that the boiling of the fuel in the fuel tank 40 is detected based on the condition of the purge concentration-related value Cpg, and the provisional detection flag Fv2a is set to a value of 1 (step S562). When the purge concentration-related value Cpg is equal to or greater than the threshold value Cpgref2, the provisional detection flag Fv2a is set to a value of 0 without determining that the boiling of the fuel in the fuel tank 40 is detected based on the condition of the purge concentration-related value Cpg (step S564). In order to suppress erroneous determination, the provisional detection flag Fv2a may be set to a value of 1 when the purge concentration-related value Cpg is less than the threshold value Cpgref2 and the duration of this is equal to or greater than a predetermined time Tpg. The predetermined time Tpg is the time required to determine that the purge concentration-related value Cpg is less than the threshold value Cpgref2, for example, several hundred msec. In addition, taking into consideration the reliability of the purge concentration related value Cpg, when the purge concentration related value Cpg is less than the threshold value Cpgref2 and the purge concentration learning count Npg is equal to or greater than the threshold value Npgref, the tentative detection flag Fv2a may be set to a value of 1.

Then, the values of the fuel pressure convergence detection flag Fpc, the integral term update continuation flag Fiu, and the consumption flow rate stability flag Ffr are checked (steps S570 to S574). The processing of steps S570 to S574 is for determining whether the feedback integral term αbi is reliable. When the fuel pressure convergence detection flag Fpc, the integral term update continuation flag Fiu, and the consumption flow rate stability flag Ffr are all set to the value 1, it is determined that the feedback integral term αbi is reliable, and the process proceeds to step S590. In other words, when fuel pressure convergence is detected, the update continuation time of the feedback integral term αbi is equal to or longer than the predetermined time Tiu, and the fuel consumption flow rate Qfr of the engine 12 is stable, the process proceeds to step S590.

Then, the difference Δβ1 is set to a value obtained by subtracting the low response processing value βlo from the high response processing value βhi (step S590), and the set difference Δβ1 is compared with the threshold value Δβref1 (step S600). Here, the threshold value Δβref1 is used to determine whether the difference Δβ1 is relatively large. Consider a case where the fuel in the fuel tank 40 is boiling and the fuel pump 44 sucks in vaporized fuel (mainly fuel vaporized by boiling) and the discharge pressure of the fuel pump 44 drops. At this time, the fuel pressure Pf becomes lower than the target fuel pressure Pf*, and the feedback integral term αbi increases (step S354 in Figures 5 to 7). Then, the greater the increase per unit time of the feedback integral term αbi at this time, the greater the difference Δβ1 obtained by subtracting the low response processing value βlo from the high response processing value βhi is likely to be. Based on this, the process of step S600 is a process of determining whether or not the boiling of the fuel in the fuel tank 40 is detected using the difference Δβ1.

When the difference Δβ1 is equal to or greater than the threshold Δβref1 in step S600, it is determined that boiling of fuel in the fuel tank 40 is detected under the condition of the difference Δβ1, and the tentative detection flag Fv2b is set to a value of 1 (step S602). When the difference Δβ1 is less than the threshold Δβref1, it is not determined that boiling of fuel in the fuel tank 40 is detected under the condition of the difference Δβ1, and the tentative detection flag Fv2b is set to a value of 0 (step S604). Note that, in order to prevent erroneous determination, the tentative detection flag Fv2b may be set to a value of 1 when the difference Δβ1 is equal to or greater than the threshold Δβref1 and its duration is equal to or greater than a predetermined time Tβ1. The predetermined time Tβ1 is the time required to determine that the difference Δβ1 is equal to or greater than the threshold Δβref1, for example, approximately several hundred msec.

The process of steps S600 to S604 will be described with reference to the explanatory diagram of FIG. 14. FIG. 14 is an explanatory diagram showing an example of the target fuel pressure Pf*, fuel pressure Pf, feedback integral term αbi, high response processing value βhi, low response processing value βlo, difference Δβ1, and tentative detection flag Fv2b. When the fuel pump 44 sucks in vaporized fuel (mainly fuel vaporized by boiling) and the discharge pressure of the fuel pump 44 drops, the feedback integral term αbi increases, and the high response processing value βhi and low response processing value βlo increase accordingly (from time t11). Then, when the difference Δβ1 obtained by subtracting the low response processing value βlo from the high response processing value βhi reaches or exceeds the threshold value Δβref1 (time t12), the tentative detection flag Fv2b is set to a value of 1. In this way, the boiling of fuel in the fuel tank 40 can be detected under the condition of the difference Δβ1.

Furthermore, the difference Δβ2 is set to a value obtained by subtracting the convergence processing value βhi0 from the current high response processing value βhi (step S610), and the set difference Δβ2 is compared with the threshold value Δβref2 (step S620). Here, the threshold value Δβref2 is used to determine whether the difference Δβ2 is relatively large. Even when the fuel in the fuel tank 40 is boiling and the fuel pump 44 sucks in vaporized fuel (mainly fuel vaporized by boiling) and the discharge pressure of the fuel pump 44 drops, if the increase in the feedback integral term αbi per unit time is not so large, the difference Δβ1 may not be equal to or greater than the threshold value Δβref1. Even in this case, if the increase in the feedback integral term αbi continues, the increase in the high response processing value βhi continues, and the difference Δβ2 obtained by subtracting the convergence processing value βhi0 from the high response processing value βhi may become relatively large. In step S620, the difference Δβ2 is determined based on this by using the difference Δβ2 to determine whether the boiling of the fuel in the fuel tank 40 is detected.

When the difference Δβ2 is equal to or greater than the threshold Δβref2 in step S620, it is determined that the boiling of fuel in the fuel tank 40 is detected based on the condition of the difference Δβ2, and the tentative detection flag Fv2c is set to a value of 1 (step S622). When the difference Δβ2 is less than the threshold Δβref2, it is not determined that the boiling of fuel in the fuel tank 40 is detected based on the condition of the difference Δβ2, and the tentative detection flag Fv2c is set to a value of 0 (step S624). Note that, in order to prevent erroneous determination, the tentative detection flag Fv2c may be set to a value of 1 when the difference Δβ2 is equal to or greater than the threshold Δβref2 and the duration of this detection is equal to or greater than a predetermined time Tβ2. The predetermined time Tβ2 is the time required to determine that the difference Δβ2 is equal to or greater than the threshold Δβref2, for example, several hundred msec.

The processing of steps S620 to S624 will be described with reference to the explanatory diagram of FIG. 15. FIG. 15 is an explanatory diagram showing an example of the target fuel pressure Pf*, fuel pressure Pf, feedback integral term αbi, high response processing value βhi, low response processing value βlo, differences Δβ1, Δβ, and provisional detection flags Fv2b and Fv2c. When the fuel pump 44 sucks in vaporized fuel (mainly fuel vaporized by boiling) and the discharge pressure of the fuel pump 44 drops, the feedback integral term αbi increases, and the high response processing value βhi and the low response processing value βlo increase accordingly (from time t21). Then, even if the difference Δβ1 obtained by subtracting the low response processing value βlo from the high response processing value βhi is less than the threshold value Δβref1, when the difference Δβ2 obtained by subtracting the convergence processing value βhi0 from the high response processing value βhi reaches or exceeds the threshold value Δβref2 (time t22), the provisional detection flag Fv2c is set to a value of 1. In this way, boiling of fuel in the fuel tank 40 can be detected based on the condition of difference Δβ2.

When the tentative detection flags Fv2a, Fv2b, and Fv2c are set in this way, the values of the tentative detection flags Fv2a, Fv2b, and Fv2c that have been set are checked (steps S630 to S634). If at least one of the tentative detection flags Fv2a, Fv2b, and Fv2c is set to a value of 1, the fuel boiling detection flag Fv is set to a value of 1 (step S632), and this routine ends. That is, when it is determined that the boiling of fuel in the fuel tank 40 is to be detected based on at least one of the conditions of the purge concentration related value Cpg (step S560), the condition of the difference Δβ1 (step S600), and the condition of the difference Δβ2 (step S620), the fuel boiling detection flag Fv is set to a value of 1. In this way, the boiling of fuel in the fuel tank 40 can be detected (changed from undetected to detected or maintained as detected).

When all of the tentative detection flags Fv2a, Fv2b, and Fv2c have a value of 0 in steps S630 to S634, the fuel boiling detection flag Fv is held at the previous value (step S634), and this routine ends. That is, when it is not determined that the boiling of fuel in the fuel tank 40 is to be detected under any of the conditions of the purge concentration related value Cpg (step S560), the difference Δβ1 (step S600), and the difference Δβ2 (step S620), the fuel boiling detection flag Fv is held at the previous value.

In steps S570 to S574, if at least one of the fuel pressure convergence detection flag Fpc, the integral term update continuation flag Fiu, and the fuel consumption flow rate stability flag Ffr is equal to 0, the process proceeds to step S580. That is, if fuel pressure convergence has not been detected, if the update duration of the feedback integral term αbi is less than the predetermined time Tiu (including when the feedback integral term αbi has not been updated), or if the fuel consumption flow rate Qfr of the engine 12 is not stable, the process proceeds to step S580.

Then, the value of the tentative detection flag Fv2a is checked (step S580). If the tentative detection flag Fv2a is equal to the value 1, the fuel boiling detection flag Fv is set to the value 1 (step S582), and this routine ends. That is, if it is determined that the boiling of fuel in the fuel tank 40 is to be detected based on the condition of the purge concentration related value Cpg (step S560), the fuel boiling detection flag Fv is set to the value 1. In this way, the boiling of fuel in the fuel tank 40 can be detected (changing from undetected to detected or maintaining detected).

When the tentative detection flag Fv2a is equal to 0 in step S580, the fuel boiling detection flag Fv is held at the previous value (step S584), and this routine ends. That is, when it is not determined that the boiling of fuel in the fuel tank 40 should be detected under the condition of the purge concentration related value Cpg (step S560), the fuel boiling detection flag Fv is held at the previous value.

Here, we will explain why the provisional detection flags Fv2b and Fv2c are not used to set the fuel boiling detection flag Fv when fuel pressure convergence has not been detected, when the update duration of the feedback integral term αbi is less than the predetermined time Tiu (including when the feedback integral term αbi has not been updated), or when the fuel consumption flow rate Qfr of the engine 12 is not stable.

When fuel pressure convergence is undetected, the fuel pressure Pf may fluctuate relatively greatly, and the feedback integral term αbi may also fluctuate relatively greatly, making it impossible to guarantee the reliability of the difference Δβ1. In addition, the difference Δβ2 cannot be calculated because the convergence processing value βhi0 cannot be set. For these reasons, in order to prevent erroneous determination of whether or not fuel boiling in the fuel tank 40 has been detected, when fuel pressure convergence is undetected, the provisional detection flags Fv2b and Fv2c are not used to set the fuel boiling detection flag Fv.

When the feedback integral term αbi is not updated, the reliability of the feedback integral term αbi cannot be guaranteed, and therefore the reliability of the differences Δβ1 and Δβ2 cannot be guaranteed. Furthermore, even if the feedback integral term αbi is updated, if the update duration is short, the feedback integral term αbi may fluctuate relatively greatly, and the reliability of the differences Δβ1 and Δβ2 cannot be guaranteed. For these reasons, in order to suppress erroneous determination of whether or not the boiling of fuel in the fuel tank 40 has been detected, when the update duration of the feedback integral term αbi is less than a predetermined time Tiu (including when the feedback integral term αbi has not been updated), the tentative detection flags Fv2b and Fv2c are not used to set the fuel boiling detection flag Fv.

When the fuel consumption flow rate Qfr of the engine 12 is not stable, the fuel pressure Pf may fluctuate relatively greatly, and the feedback proportional term αbp and the feedback integral term αbi may also fluctuate relatively greatly, making it impossible to guarantee the reliability of the differences Δβ1 and Δβ2. For this reason, in order to prevent erroneous determination of whether or not the boiling of fuel in the fuel tank 40 has been detected, when the fuel consumption flow rate Qfr of the engine 12 is not stable, the provisional detection flags Fv2b and Fv2c are not used to set the fuel boiling detection flag Fv.

In the engine device 11 equipped in the automobile 10 of the embodiment described above, when boiling in the fuel tank 40 is detected, the target rotation speed Np* of the fuel pump 44 is set within a range of a lower limit rotation speed Npmin as the lower limit of the rotation speed range of the fuel pump 44 at which the fuel pump 44 can discharge the vaporized fuel (mainly fuel vaporized by boiling) from the discharge hole 44a when the fuel pump 44 sucks in the vaporized fuel (mainly fuel vaporized by boiling) and an upper limit rotation speed Npmax as the upper limit of the rotation speed range of the fuel pump 44 at which the generation of vaporized fuel in the fuel pump 44 due to the heat generation of the fuel pump 44 can be suppressed, and the fuel pump 44 is controlled. As a result, even if the fuel pump 44 sucks in the vaporized fuel when the fuel in the fuel tank 40 is boiling, the vaporized fuel can be discharged and the generation of vaporized fuel in the fuel pump due to the heat generation of the fuel pump 44 can be suppressed.

In the engine device 11 of the embodiment, in the fuel pump control routine of Figs. 5 to 7, when the fuel boiling detection flag Fv is set to a value of 1 in step S320, the relief valve 46 opening pressure Pfrv is set to the target fuel pressure Pf* in step S380. However, this is not limited to this, and it is sufficient that when the fuel boiling detection flag Fv is set to a value of 1, the target fuel pressure Pf* is set to a value higher than when the flag is set to a value of 0.

In the engine device 11 of the embodiment, in the fuel pump control routine of FIG. 5 to FIG. 7, when the fuel boiling detection flag Fv is value 1 in step S320, that is, when the boiling pump control is executed, the integrated execution time Tfup of the boiling pump control is measured in step S408. That is, when the boiling pump control is executed, the integrated execution time Tfup is measured uniformly regardless of the factor caused by the execution. However, the integrated execution time Tfup[i] (i: variable for the factor) may be measured for each factor caused by the execution of the boiling pump control. Examples of each factor include the provisional undetected flag Fv1a being value 1 (the purge concentration related value Cpg is less than the threshold value Cpgref2), the provisional detection flag Fv2b and/or the provisional detection flag Fv2c (the difference Δβ1 is equal to or greater than the threshold value Δβref1 and/or the difference Δβ2 is equal to or greater than the threshold value Δβref2), etc. By doing this, manufacturers, dealers, and others will be able to carry out various analyses more appropriately.

In the engine device 11 of the embodiment, the electronic control unit 70 executes the fuel pump control routine of Figs. 5 to 7 to control the fuel pump 44. However, instead of this, the electronic control unit 70 may execute the fuel pump control routine of Figs. 5 to 7 by replacing the part of Fig. 6 with the part of Fig. 16. The part of the fuel pump control routine of Fig. 16 differs from the part of the fuel pump control routine of Fig. 6 in that the processes of steps S430 to S434 are added.

In a part of the fuel pump control routine in FIG. 16, when the accumulated execution time Tfup of the boiling pump control is timed in step S408, the convergence processing value βhi0 is input (step S432) on the condition that the convergence processing value βhi0 can be input (step S430), the input convergence processing value βhi0 is set to the feedback integral term αbi (step S434), and this routine ends. Here, being able to input the convergence processing value βhi0 means that the convergence processing value βhi0 has been set and not reset in step S516 of the fuel boiling detection flag setting routine in FIG. 10 to FIG. 12.

The processing of steps S430 to S434 will be described with reference to the explanatory diagram of Figure 17. Figure 17 is an explanatory diagram showing an example of the target fuel pressure Pf*, fuel pressure Pf, feedback integral term αbi, high response processing value βhi, fuel pressure convergence detection flag Fpc, and fuel boiling detection flag Fv. When the condition that the fuel pressure difference ΔPf is equal to or less than the threshold value ΔPfref and the duration of this is equal to or more than the predetermined time Tpf is met (time t31), the fuel pressure convergence detection flag Fpc is switched from value 0 to value 1, and the high response processing value βhi at that time is set to the convergence processing value βhi0. After that, the fuel pump 44 sucks in vaporized fuel (mainly fuel vaporized by boiling) and the discharge pressure of the fuel pump 44 drops, and the feedback integral term αbi and the high response processing value βhi increase accordingly. When the fuel boiling detection flag Fv switches from value 0 to value 1 due to the above-mentioned purge concentration related value Cpg, difference Δβ1, and difference Δβ2 (time t32), the target fuel pressure Pf* is increased to the opening pressure Pfrv of the relief valve 46 and the high response processing value βhi is set to the convergence processing value βhi0.

It is assumed that the feedback integral term αbi set in step S354 immediately before the fuel boiling detection flag Fv switches from value 0 to value 1 includes the influence of the fuel boiling in the fuel tank 40. Therefore, if the processing of step S434 is not performed and the feedback integral term αbi is set (updated) in step S354 after the fuel boiling detection flag Fv switches from value 1 to value 0, the feedback integral term αbi may become a value that includes the influence when the fuel in the fuel tank 40 is boiling. In light of this, in this modified example, when the fuel boiling detection flag Fv is value 1, the convergence processing value βhi0 is set to the feedback integral term αbi. This makes it possible to set the feedback integral term αbi to a more appropriate value when the feedback integral term αbi is set (updated) in step S354 after the fuel boiling detection flag Fv switches from value 1 to value 0.

In the fuel pump control routine of FIG. 16, when the fuel boiling detection flag Fv is set to the value 1 in step S320, the convergence processing value βhi0 is set to the feedback integral term αbi in step S434. However, instead of this, when the fuel boiling detection flag Fv is set to the value 1, the convergence feedback integral term αbi0, which is the feedback integral term αbi immediately after the fuel pressure convergence detection flag Fpc switches from the value 0 to the value 1 (immediately after the fuel pressure convergence is changed from undetected to detected), may be set to the feedback integral term αbi.

In the engine device 11 of the embodiment, in the fuel pump control routine of Figures 5 to 7, when the engine revolution flag Fer is set to a value of 1 in step S310, the fuel boiling detection flag Fv is set to a value of 0 in step S320, and the fuel cut flag Ffc1 is set to a value of 0 in step S340, the feedback integral term αbi is set in step S354. However, instead of this, when the engine revolution flag Fer is set to a value of 1, the fuel boiling detection flag Fv is set to a value of 0, and the fuel cut flag Ffc1 is set to a value of 0, the feedback integral term αbi may be set by the feedback integral term setting process of Figure 18.

In the feedback integral term setting process of FIG. 18, the electronic control unit 70 first sets the target fuel pressure change amount δPf* to the absolute value of the difference between the current value of the target fuel pressure Pf* and the previous value (step S800), and sets the fuel pressure difference ΔPf2 to the absolute value of the difference between the target fuel pressure Pf* and the fuel pressure Pf (step S810). Next, the target fuel pressure change amount δPf* is compared with a threshold value δPfref (step S820). Here, the threshold value δPfref is used to determine whether the target fuel pressure Pf* has changed relatively significantly. For this threshold value δPfref, a value smaller than the target fuel pressure change amount δPf* when the fuel boiling detection flag Fv switches from value 1 to value 0 is used.

When the target fuel pressure change amount δPf* is less than the threshold value δPfref in step S820, it is determined that the target fuel pressure change amount δPf* has not changed significantly, and the value of the large change flag Fch is checked (step S840). Here, the large change flag Fch is set to a value of 1 when the target fuel pressure change amount δPf* becomes equal to or greater than the threshold value δPFref by this routine, and is set to a value of 0 when the release condition is subsequently satisfied. For example, the large change flag Fch is set to a value of 0 as an initial value at the start of a trip. When the large change flag Fch is set to a value of 0 in step S840, the feedback integral term αbi is set by the above-mentioned formula (2) in the same manner as in the processing of step S354 of the fuel pump control routine of Figures 5 to 7 (step S880), and the feedback integral term setting processing is terminated.

When the target fuel pressure change amount δPf* is equal to or greater than the threshold value δPfref in step S820, it is determined that the target fuel pressure Pf* has changed relatively significantly, and the large change flag Fch is set to a value of 1 (step S822). The elapsed time Tch since the target fuel pressure change amount δPf* became equal to or greater than the threshold value δPfref is reset to a value of 0 and timing is then started (step S824).

Next, the cause of the target fuel pressure change amount δPf* becoming equal to or greater than the threshold value δPfref is examined (step S826). When the cause of the target fuel pressure change amount δPf* becoming equal to or greater than the threshold value δPfref is not the fuel boiling detection flag Fv switching from value 1 to value 0, a predetermined time Tch1 is set as the upper limit time Tchlim for holding the large change flag Fch at value 1 (step S828). On the other hand, when the cause of the target fuel pressure change amount δPf* becoming equal to or greater than the threshold value δPfref is the fuel boiling detection flag Fv switching from value 1 to value 0, a predetermined time Tch2 shorter than the predetermined time Tch1 is set as the upper limit time Tchlim (step S830). Here, the predetermined time Tch1 is approximately several thousand to several tens of thousands of msec, and the predetermined time Tch1 is approximately 1/3 to 2/3 of the predetermined time Tch1. Then, the feedback integral term αbi is held at the previous value (step S860), and the feedback integral term setting process is terminated. If the feedback integral term αbi is updated when the target fuel pressure change amount δPf* is relatively large, the absolute value of the feedback integral term αbi becomes large, and there is a possibility that the fuel pressure Pf will overshoot or undershoot the target fuel pressure Pf*. For this reason, when the target fuel pressure change amount δPf* is equal to or greater than the threshold value δPfref, the feedback integral term αbi is held at the previous value.

When the target fuel pressure change amount δPf* is less than the threshold value δPfref in step S820 and the large change flag Fch is set to a value of 1 in step S840, the elapsed time Tch is compared with the upper limit time Tchlim (step S850). When the elapsed time Tch is less than the upper limit time Tchlim, the elapsed time Tch is compared with the threshold value Tchref (step S852), and the fuel pressure difference ΔPf2 is compared with the threshold value ΔPfref2 (step S854). Here, the predetermined time Tchref is the time during which the fuel pressure Pf can reach the vicinity of the target fuel pressure Pf* after the target fuel pressure change amount δPf* becomes equal to or greater than the threshold value δPfref. The threshold value ΔPfref2 is used to determine whether the fuel pressure Pf has reached the vicinity of the target fuel pressure Pf* after the target fuel pressure change amount δPf* becomes equal to or greater than the threshold value δPfref.

In steps S852 and S854, if the elapsed time Tch is less than the threshold value Tchref or if the fuel pressure difference ΔPf2 is greater than the threshold value ΔPfref2, it is determined that the time during which the fuel pressure Pf can reach the vicinity of the target fuel pressure Pf* has not elapsed after the target fuel pressure change amount δPf* becomes equal to or greater than the threshold value δPfref and/or that the fuel pressure Pf has not reached the vicinity of the target fuel pressure Pf*, the feedback integral term αbi is held at the previous value (step S860), and the feedback integral term setting process is terminated.

In steps S852 and S854, when the elapsed time Tch is equal to or greater than the threshold value Tchref and the fuel pressure difference ΔPf2 is equal to or less than the threshold value ΔPfref2, it is determined that the time during which the fuel pressure Pf can reach the vicinity of the target fuel pressure Pf* has elapsed after the target fuel pressure change amount δPf* becomes equal to or greater than the threshold value δPfref and that the fuel pressure Pf has reached the vicinity of the target fuel pressure Pf*, the large change flag Fch is set to a value of 0 (step S870), the feedback integral term αbi is set by the processing of step S880 described above, and the feedback integral term setting processing is terminated.

When the elapsed time Tch is equal to or greater than the upper limit time Tchlim in step S850, the large change flag Fch is set to 0 (step S870), the feedback integral term αbi is set by the processing of step S880 described above, and the feedback integral term setting processing is terminated. This makes it possible to prevent the time until the fuel pressure Pf reaches the vicinity of the target fuel pressure Pf* from becoming excessively long. Moreover, in this modified example, when the cause of the target fuel pressure change amount δPf* becoming equal to or greater than the threshold value δPfref is that the fuel boiling detection flag Fv has switched from value 1 to value 0, the upper limit time Tchlim is made shorter than in other cases, thereby making it possible to further prevent the time until the fuel pressure Pf reaches the vicinity of the target fuel pressure Pf* from becoming long.

In the engine device 11 of the embodiment, in the fuel pump control routine of Figures 5 to 7, when the engine rotation flag Fer is set to a value of 0 in step S310, the fuel pump 44 is set to a rotation stop state. However, regardless of the value of the engine rotation flag Fer, normal pump control or boiling pump control may be executed based on the fuel boiling detection flag Fv.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, a constant value is used as the threshold value Cpgref1 used for comparison with the purge concentration related value Cpg in step S522. However, the threshold value Cpgref1 may be set based on the atmospheric pressure Pout. In this case, the threshold value Cpgref1 can be set, for example, by applying the atmospheric pressure Pout to a first purge concentration threshold setting map. The first purge concentration threshold setting map is predetermined as a relationship between the atmospheric pressure Pout and the threshold value Cpgref1, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 19 is an explanatory diagram showing an example of the first purge concentration threshold setting map. As shown in the figure, the threshold value Cpgref1 is set to be larger (the absolute value becomes smaller) as the atmospheric pressure Pout is lower within a negative range. This is because, based on the fact that the lower the atmospheric pressure Pout, the more likely it is that the absolute pressure in the fuel tank 40 will fall below the saturated vapor pressure of the fuel, and the more likely the fuel in the fuel tank 40 will boil, the lower the atmospheric pressure Pout, the more difficult it is to leave the boiling of the fuel in the fuel tank 40 undetected.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, a constant value is used as the threshold value Taref used for comparison with the intake air temperature Ta in step S540. However, the threshold value Taref may be set based on the atmospheric pressure Pout. In this case, the threshold value Taref can be set, for example, by applying the atmospheric pressure Pout to an intake air temperature threshold setting map. The intake air temperature threshold setting map is determined in advance as the relationship between the atmospheric pressure Pout and the threshold value Taref, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 20 is an explanatory diagram showing an example of the intake air temperature threshold setting map. As shown in the figure, the threshold value Taref is set to be lower as the atmospheric pressure Pout is lower. This is based on the same reason as the tendency of the first purge concentration threshold setting map.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of Figures 10 to 12, the provisional undetected flag Fv1c is set using the intake air temperature Ta in step S540. However, instead of the intake air temperature Ta, the provisional undetected flag Fv1c may be set using the fuel temperature Tf, which is the temperature of the fuel in the fuel tank 40. Here, the fuel temperature Tf may be estimated based on the intake air temperature Ta, or may be detected by a temperature sensor attached to the fuel tank 40.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, a constant value is used as the threshold value Cpgref2 used for comparison with the purge concentration related value Cpg in step S560. However, the threshold value Cpgref2 may be set based on the atmospheric pressure Pout. In this case, the threshold value Cpgref2 can be set, for example, by applying the atmospheric pressure Pout to a second purge concentration threshold setting map. The second purge concentration threshold setting map is predetermined as a relationship between the atmospheric pressure Pout and the threshold value Cpgref2, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 21 is an explanatory diagram showing an example of the second purge concentration threshold setting map. As shown in the figure, the threshold value Cpgref2 is set to be larger (the absolute value becomes smaller) as the atmospheric pressure Pout is lower within a negative range. This is because, considering that the lower the atmospheric pressure Pout, the more likely it is that the absolute pressure in the fuel tank 40 will fall below the saturated vapor pressure of the fuel, and the more likely the fuel in the fuel tank 40 will boil, the lower the atmospheric pressure Pout, making it easier to detect the boiling of the fuel in the fuel tank 40.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, a constant value is used as the threshold value Δβref1 used for comparison with the difference Δβ1 in step S600. However, the threshold value Δβref1 may be set based on the atmospheric pressure Pout. In this case, the threshold value Δβref1 can be set, for example, by applying the atmospheric pressure Pout to the first difference threshold setting map. The first difference threshold setting map is determined in advance as the relationship between the atmospheric pressure Pout and the threshold value Δβref1, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 22 is an explanatory diagram showing an example of the first difference threshold setting map. As shown in the figure, the threshold value Δβref1 is set to be smaller within a positive range as the atmospheric pressure Pout is lower. This is based on the same reason as the tendency of the second purge concentration threshold setting map.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, a constant value is used as the threshold value Δβref2 used for comparison with the difference Δβ2 in step S620. However, the threshold value Δβref2 may be set based on the atmospheric pressure Pout. In this case, the threshold value Δβref2 can be set, for example, by applying the atmospheric pressure Pout to the second difference threshold setting map. The second difference threshold setting map is determined in advance as the relationship between the atmospheric pressure Pout and the threshold value Δβref2, and is stored in the ROM or flash memory of the electronic control unit 70. FIG. 23 is an explanatory diagram showing an example of the second difference threshold setting map. As shown in the figure, the threshold value Δβref2 is set to be smaller within a positive range as the atmospheric pressure Pout is lower. This is based on the same reason as the tendency of the second purge concentration threshold setting map.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, the difference Δβ1 is set to the value obtained by subtracting the low response processing value βlo from the high response processing value βhi in step S590. However, the difference Δβ1 may also be set to the value obtained by subtracting the low response processing value βlo from the feedback integral term αbi.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of Figures 10 to 12, the difference Δβ2 is set to the value obtained by subtracting the convergence time processing value βhi0 from the current high response processing value βhi in step S610. However, the feedback integral term αbi immediately after the fuel pressure convergence detection flag Fpc switches from value 0 to value 1 may be stored as the convergence time feedback integral term αbi0, and the difference Δβ2 may be set to the value obtained by subtracting the convergence time feedback integral term αbi0 from the current feedback integral term αbi.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, the process of determining whether to set the fuel boiling detection flag Fv to a value of 0 (steps S550 to S554) uses a provisional undetected flag Fv1a based on whether the conditions of the purge concentration learning count Npg, the purge concentration related value Cpg, and the tank internal pressure Pt are satisfied, a provisional undetected flag Fv1b based on whether the conditions of the cooling water temperature Tw are satisfied, and a provisional undetected flag Fv1c based on whether the conditions of the intake air temperature Ta are satisfied. However, only some of the provisional undetected flags Fv1a, Fv1b, and Fv1c may be used in this determination process.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of FIG. 10 to FIG. 12, the process of determining whether to set the fuel boiling detection flag Fv to a value of 1 (steps S630 to S634) uses a tentative detection flag Fv2a based on whether the condition of the purge concentration related value Cpg is satisfied, a tentative detection flag Fv2b based on whether the condition of the difference Δβ1 is satisfied, and a tentative detection flag Fv2c based on whether the condition of the difference Δβ2 is satisfied. However, only some of the tentative detection flags Fv2a, Fv2b, and Fv2c may be used in this determination process.

In the engine device 11 of the embodiment, in the fuel boiling detection flag setting routine of Figs. 10 to 12, the fuel pressure convergence detection flag Fpc, the integral term update continuation flag Fiu, and the consumption flow rate stable flag Ffr are used in the process of determining whether or not to use the tentative detection flags Fv2b, Fv2c to set the fuel boiling detection flag Fv (steps S570 to S574). However, in this determination process, it is sufficient to use at least the fuel pressure convergence detection flag Fpc, and at least one of the integral term update continuation flag Fiu and the consumption flow rate stable flag Ffr may not be used.

In the engine device 11 of the embodiment, in the fuel pressure convergence detection flag setting routine of FIG. 13, the absolute value of the value obtained by subtracting the pulsation center fuel pressure Pfcn from the target fuel pressure Pf* is set to the fuel pressure difference ΔPf in step S730. However, instead of this, the absolute value of the value obtained by subtracting the fuel pressure Pf from the target fuel pressure Pf* may be set to the fuel pressure difference ΔPf.

In the engine device 11 of the embodiment, in the fuel pressure convergence detection flag setting routine of FIG. 13, the fuel pump rotation flag Fpr, the target fuel pressure Pf*, and the fuel cut continuation flag Ffc2 are used in the process of determining whether or not to set the fuel pressure convergence detection flag Fpc to a value of 0 (steps S710 to S714). However, only some of the fuel pump rotation flag Fpr, the target fuel pressure Pf*, and the fuel cut continuation flag Ffc2 may be used in this determination process.

In the engine device 11 equipped in the automobile 10 of the embodiment, the engine 12 is equipped with a fuel injection valve 26 (in-cylinder injection valve) that injects fuel into a cylinder. However, instead of or in addition to the fuel injection valve 26, a port injection valve that injects fuel into an intake port may be provided.

In the embodiment, the engine device 11 is provided in an automobile 10 that runs using power from an engine 12. However, the engine device 11 may be provided in any form provided in a vehicle that can operate the engine 12 intermittently, and may be provided in the form of an engine device 11 provided in a hybrid automobile that has a motor in addition to an engine.

The following explains the relationship between the main elements of the embodiment and the main elements of the invention described in the section on means for solving the problem. In the embodiment, the engine 12 corresponds to the "engine", the fuel supply device 42 corresponds to the "fuel supply device", and the electronic control unit 70 corresponds to the "control device".

The correspondence between the main elements of the Examples and the main elements of the invention described in the Means for Solving the Problem column does not limit the elements of the invention described in the Means for Solving the Problem column, since the Examples are examples for specifically explaining the form for implementing the invention described in the Means for Solving the Problem column. In other words, the interpretation of the invention described in the Means for Solving the Problem column should be based on the description in that column, and the Examples are merely a specific example of the invention described in the Means for Solving the Problem column.

The above describes the form for carrying out the present invention using examples, but the present invention is not limited to these examples in any way, and it goes without saying that the present invention can be carried out in various forms without departing from the spirit of the present invention.

The present invention can be used in the engine equipment manufacturing industry, etc.

10 automobile, 11 engine device, 12 engine, 14 crankshaft, 14a crank position sensor, 15 water temperature sensor, 16 cam position sensor, 22 air cleaner, 23 intake pipe, 23a air flow meter, 23b temperature sensor, 24 throttle valve, 24a throttle position sensor, 26 fuel injection valve, 28 intake valve, 29 combustion chamber, 30 spark plug, 32 piston, 33 exhaust valve, 34 exhaust pipe, 35 purification device, 37 front air-fuel ratio sensor, 38 rear air-fuel ratio sensor, 40 fuel tank, 40a pressure sensor, 42 fuel supply device, 43 fuel flow path, 43p fuel pressure sensor, 44 fuel pump, 44a exhaust hole, 44b rotational speed sensor, 45 relief passage, 46 relief valve, 50 vaporized fuel treatment device, 51 introduction passage, 52 Shut-off valve, 52a Shut-off valve position sensor, 53 Canister, 54 Atmospheric release passage, 55 Purge passage, 56 Purge valve, 56a Purge valve position sensor, 60 Transmission, 62 Differential gear, 64a, 64b Drive wheels, 70 Electronic control unit, 80 Ignition switch, 81 Shift lever, 82 Shift position sensor, 83 Accelerator pedal, 84 Accelerator pedal position sensor, 85 Brake pedal, 86 Brake pedal position sensor, 88 Vehicle speed sensor, 89 Atmospheric pressure sensor.

Claims (1)

The engine,
a fuel supply device having a fuel pump that supplies fuel in a fuel tank to a fuel passage connected to a fuel injection valve of the engine;
A control device for controlling the fuel pump;
An engine device comprising:
when the control device detects boiling of the fuel in the fuel tank, the control device sets a target rotation speed of the fuel pump within a range including a lower limit rotation speed at which the fuel pump can discharge the vaporized fuel when the vaporized fuel is sucked in by the fuel pump and an upper limit rotation speed at which generation of the vaporized fuel in the fuel pump due to an amount of heat generated by the fuel pump can be suppressed , and controls the fuel pump;
The control device includes:
When boiling of the fuel in the fuel tank is not detected, the target rotation speed is set by using a feedback control of a proportional term and an integral term based on a difference between a fuel pressure, which is a pressure of the fuel in the fuel flow path, and a target fuel pressure, to control the fuel pump;
a second processing value obtained by subjecting the integral term to a smoothing process with a second response lower than the first response from a first processing value obtained by subjecting the integral term to a smoothing process with a first response to calculate a difference, and when the difference is equal to or greater than a difference threshold value which decreases as the atmospheric pressure decreases, boiling of fuel in the fuel tank is detected.
Engine equipment.

Images