JP2007239549A - Vehicle control method and control device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To enable highly-accurate measurement of the atmospheric pressure without being influenced by pressure loss of a canister. <P>SOLUTION: A vehicle has a function of automatically stopping an engine when a predetermined operation condition is satisfied and automatically restarting the engine when another predetermined operation condition is satisfied. The engine comprises the canister (54), a purge passage (56), a purge control valve (61), a drain cut valve (62) and a pressure detection means (63) and includes a leak diagnosis procedure for diagnosing whether or not there is leak with use of the purge control valve (61), the drain cut valve (62) and the pressure detection means (63) when a leak diagnosis enabling condition is satisfied, a valve opening procedure for opening the purge control valve (61) in automatic stopping of the engine and a procedure for taking in flow passage pressure detected by the pressure detection means (63) as an atmospheric pressure measurement in the engine automatic stop when the purge control valve (61) is opened under the valve opening procedure. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to a vehicle control method and control device, and more particularly to a device for measuring atmospheric pressure using a pressure sensor used for leak diagnosis.

In an engine (internal combustion engine) that includes an evaporative fuel processing device and performs a leak diagnosis, since a new sensor for detecting the atmospheric pressure necessary for engine control increases the cost, the pressure sensor used for the leak diagnosis increases the cost. There is one that measures atmospheric pressure (see Patent Document 1). This is because when the purge control valve is fully closed during engine operation, the flow path from the canister's atmosphere opening to the purge control valve is opened to the atmosphere via the atmosphere opening. In the fully closed state, the flow path pressure detected by the pressure sensor is taken in as an atmospheric pressure measurement value.
JP 2003-35216 A

  However, this same engine that has an evaporative fuel processing device and performs leak diagnosis, particularly when used in a so-called hybrid vehicle that drives a vehicle using at least one of a motor and an engine, is affected by the pressure loss of the canister. It has been newly found that there is a situation where atmospheric pressure cannot be measured easily and accurately. That is, in order to measure the atmospheric pressure with the pressure sensor, it is necessary to capture the opportunity for the pergin control valve to be fully closed. However, an engine used for a hybrid vehicle has fewer opportunities to operate the engine while the vehicle is running, unlike a vehicle using only the engine as a power source. For this reason, in an engine used in a hybrid vehicle, the purge control valve can be said to be left open to purge fuel vapor (gasoline particles) from the canister during engine operation. There are few opportunities to come. Nevertheless, when the atmospheric pressure is to be measured, the purge control valve is not fully closed, that is, the pressure at the low purge flow rate is taken in as the atmospheric pressure measurement value. However, being at a low purge flow rate (during purge) means that pressure loss occurs in the canister if gasoline particles are adsorbed on the activated carbon, so the pressure monitored at the low purge flow rate is atmospheric pressure. If the measurement value is used, an error corresponding to the pressure loss of the canister occurs in the atmospheric pressure measurement value.

  On the other hand, the hybrid vehicle has a function of automatically stopping the engine when a predetermined operating condition is satisfied, and automatically restarting the engine when another predetermined operating condition is satisfied. At the time of automatic stop, the purge control valve is fully closed. For this reason, it can be considered that the pressure monitored when the engine is automatically stopped is used as an atmospheric pressure measurement value.However, when gasoline particles are adsorbed on the activated carbon and a pressure loss occurs in the canister, it is because the engine is automatically stopped. In other words, the pressure of the flow path from the canister's atmosphere opening to the purge control valve does not immediately become atmospheric pressure, but the pressure of the flow path from the canister's atmosphere opening to the purge control valve becomes atmospheric pressure with a delay in response. Converge. Therefore, this response delay time is set as a delay time, and the pressure after the delay time has elapsed after the engine is automatically stopped is taken in as an atmospheric pressure measurement value, so that an accurate atmospheric pressure measurement value can be obtained. Can be considered.

  However, in this method, the more the gasoline particles adsorbed on the activated carbon, the longer the response delay time until the pressure of the flow path from the canister air opening to the purge control valve converges to atmospheric pressure. If a delay time shorter than the longer response delay time is set, an error occurs in the atmospheric pressure measurement value. However, if the delay time is increased, it may occur that the atmospheric pressure measurement value cannot be captured when the engine automatic stop time is short.

  As described above, as long as the method of Patent Document 1 in which air is introduced into the flow path to the purge control valve via the canister, the influence of the pressure loss of the canister is particularly affected in a hybrid vehicle where the engine is unlikely to be operated. As a result, the atmospheric pressure cannot be measured easily and accurately.

  Accordingly, the present invention provides a vehicle control method and control device that can easily and accurately measure atmospheric pressure without being affected by canister pressure loss even in a hybrid vehicle in which the engine is operated less frequently. The purpose is to provide.

  The present invention provides a vehicle having a function of automatically stopping an engine when a predetermined operating condition is satisfied, and automatically restarting the engine when another predetermined operating condition is satisfied. A canister that guides and adsorbs the generated fuel vapor; a purge passage that communicates the canister and an intake pipe downstream of the throttle valve; a purge control valve that opens and closes the purge passage; and a drain that opens and closes the atmosphere opening of the canister A cut valve and a pressure detection means for detecting the pressure of the flow path from the fuel tank to the purge control valve. When a leak diagnosis permission condition is satisfied, the purge control valve, the drain cut valve, and the pressure detection means are While using it to diagnose whether there is a leak, To open the purge control valve, when opening the purge control valve constitutes a pressure is being passage detected by the pressure detecting means to capture the engine automatic stop atmospheric pressure measurements.

  There is a canister that causes pressure loss in the flow path from the purge control valve to the atmosphere opening of the canister, whereas the flow path from the purge control valve to the air cleaner upstream of the intake pipe Therefore, when the engine is automatically stopped, the pressure in the flow path from the purge control valve to the air cleaner upstream of the intake pipe is atmospheric pressure. According to the present invention, when the engine is automatically stopped, the purge control valve is opened, and the atmospheric pressure existing in the flow path from the purge control valve to the air cleaner upstream of the intake pipe is introduced, so that the pressure detection means immediately contacts the atmospheric pressure. Thus, the atmospheric pressure can be obtained easily and accurately. Every time the engine is automatically stopped, the atmospheric pressure measurement value with less error can be acquired. In other words, the purge passage not only communicates with the atmosphere via the air opening of the canister, but also communicates with the atmosphere via the intake pipe, so the atmosphere is introduced from the intake pipe. By doing so, it is possible to easily obtain an atmospheric pressure with little error regardless of the pressure loss of the canister.

  Embodiments of the present invention will be described with reference to the drawings.

  FIG. 1 is a schematic configuration diagram of a vehicle control apparatus used directly for carrying out the vehicle control method, and FIG. 2 is a schematic configuration diagram of a control system of the vehicle.

  The illustrated vehicle is a so-called hybrid vehicle in which the vehicle is driven using at least one of a motor and an engine. The present invention has an engine that includes an evaporative fuel processing device and performs a leak diagnosis. It is characterized in that it is applied to a hybrid vehicle having a function of automatically stopping the engine and restarting the engine automatically when another predetermined driving condition is satisfied. Since there is not, there will be an outline of the configuration of the hybrid vehicle.

  1 and 2, reference numeral 2 denotes an engine, 4 denotes a continuously variable automatic transmission, and a motor generator 3 is disposed between them. The rotation of the engine 2 or the motor generator 3 is transmitted from the continuously variable transmission 4 to the drive wheel 7 (rear wheel) via the drive shaft 5 and the differential gear 6.

  The continuously variable automatic transmission 4 is composed of, for example, a torque converter, a forward / reverse switching mechanism, and a metal belt wound around a variable pulley, and the speed transmitted through the metal belt by changing the pulley ratio of the variable pulley. The ratio changes. The primary hydraulic pressure for driving the variable pulley is set so that the target gear ratio of the continuously variable automatic transmission 4 is set according to the operating state, and this matches the gear ratio that is the ratio of the actual input rotation speed to the output rotation speed. And the secondary hydraulic pressure are controlled.

  The forward / reverse switching mechanism reverses the direction of output rotation between forward and reverse, and the torque converter transmits the input rotational torque to the output side via fluid force, and outputs at the time of extremely low speed rotation on the input side. The rotation of the side can be allowed to stop.

  The motor generator 3 is directly connected to the crankshaft of the engine 2 or connected via a belt or chain, and rotates in synchronization with the engine 2. The motor generator 3 functions as an electric motor or a generator. When the motor generator 3 functions as a motor supplementing the output of the engine 2 or as a motor for starting the engine 2, a current from the battery (42V battery) 8 is supplied via the inverter 9, and the vehicle When functioning as a generator to recover travel energy, the battery 8 is charged by the current generated through the inverter 9.

  On the other hand, another motor generator 11 is provided, and the rotation of the motor generator 11 is transmitted to the drive wheel 15 (front wheel) via the reduction gear 12, the drive shaft 13, and the differential gear 14. The motor generator 11 also functions as an electric motor or a generator. When the motor generator 11 also functions as an electric motor, the current from the battery 8 is supplied via the inverter 16. When the motor generator 11 functions as a generator for recovering the running energy of the vehicle, the battery 8 is generated by the current generated via the inverter 16. Is charged.

  Hereinafter, the motor generators 3 and 11 are simply referred to as “motors”.

  For this reason, signals from the accelerator sensor 31 and the vehicle speed sensor are input to the hybrid controller 21 (see FIG. 2). The hybrid controller 21 receives the engine controller 22, the transmission controller 23, the battery controller 24, and the motor controller 25 based on these signals. Controls during acceleration, constant speed, and deceleration while cooperating. Actually, no vehicle speed sensor is provided, and the vehicle speed is calculated based on the engine rotation speed detected by the engine rotation speed sensor 32, the gear ratio of the transmission 4, and the like.

  Here, if the driving torque is separately transmitted to the front wheel 15 and the rear wheel 7, 4WD traveling becomes possible. Therefore, when the 4WD switch 33 provided in the vehicle interior is turned on, the hybrid controller 21 performs the creep traveling state. The vehicle is started from 4WD.

  In addition, an assist switch 34 is provided in the passenger compartment so that a predetermined acceleration feeling can be obtained when necessary. When the driver turns on the assist switch 34, the hybrid controller 21 assists the driving force by the motor 11.

  On the other hand, the engine 2 is automatically stopped (idle stop) when a predetermined driving condition (idle stop permission condition) is satisfied while the vehicle is traveling, and then another predetermined driving condition is satisfied (idle stop). The hybrid controller 21 stops the operation of the engine 2 when a predetermined operating condition is satisfied during the traveling of the vehicle, and then after that, the engine 2 is automatically restarted when the permission condition is no longer satisfied. The engine 2 is started by the motor 3 when another predetermined operating condition is satisfied. As can be seen from the conditions [1] to [5] described later, the idle stop permission condition does not include the condition that the vehicle speed = 0 km / h and the brake is operating. That is, the engine is automatically stopped mainly while the vehicle is traveling, and restart after the engine is automatically stopped is also performed while the vehicle is traveling.

  Therefore, in addition to the accelerator sensor 31 and the engine rotation speed sensor 32, the hybrid controller 21 receives signals from the shift position sensor 36, the intake pressure sensor 38, the steering angle sensor 39, etc. of the continuously variable transmission 4, Based on these, automatic stop and restart of the engine 2 are controlled via the engine controller 22.

  During operation of the engine 2, the engine controller 22 controls the opening of the throttle valve 42 according to the accelerator opening and the engine speed, and controls the fuel injection amount from the fuel injection valve 43 and the timing of fuel injection. In addition, the ignition timing, which is the timing at which the spark plug 44 ignites sparks, is controlled to generate the engine output that provides the required driving force. When the hybrid controller 21 receives a command to automatically stop the engine, Then, when the operation of the fuel injection valve 43 and the spark plug 44 is stopped, and then the engine restart command is received from the hybrid controller 21, the operation of the fuel injection valve 43 and the spark plug 44 is resumed.

  FIG. 3 shows a schematic configuration diagram of a control system of the evaporated fuel processing apparatus provided in the engine 2.

  In FIG. 3, the evaporated fuel processing apparatus mainly includes a first passage 52, a canister 54, a second passage 56 (purge passage), and a purge control valve 61.

  The fuel vapor (vapor) in the upper part of the fuel tank 51 is guided to the canister 54 via the first passage 52, and only the fuel particles (gasoline particles) are adsorbed by the activated carbon 54 a in the canister 54, and the remaining air is passed through the canister 54. It is discharged to the outside through an air opening 55 provided in the vertical lower part (shown in the upper part of the canister 54 in the figure).

  The canister 54 communicates with an intake pipe 57 downstream of the throttle valve 42 through a second passage 56, and a normally closed purge control valve 61 driven by a step motor is provided in the second passage 56. When the purge control valve 61 is opened in response to a signal from the engine controller 22 under a certain condition (for example, a low load region after the engine warm-up is completed), the pressure that is greatly developed downstream of the throttle valve 42 and less than atmospheric pressure Fresh air is introduced into the canister 54 from the atmosphere opening 55 of the canister 54. With this fresh air, gasoline particles are introduced from the activated carbon 54 a together with fresh air into the intake pipe 57 through the second passage 56 and burned in the combustion chamber together with the fuel injected from the fuel injection valve 43.

  Since the second passage 56 includes a passage 56a from the canister 54 to the purge control valve 61 and a passage 56b from the purge control valve 61 to the outlet of the intake pipe 57, hereinafter, from the canister 54 to the purge control valve 61, The passage is referred to as “canister side second passage” 56a, and the passage from the purge control valve 61 to the intake pipe 57 is referred to as “intake pipe side second passage” 56b.

  On the other hand, a normally open drain cut valve 62 is provided at the atmosphere opening 55 of the canister 54. This drain cut valve 62 is closed at the time of leak diagnosis, and is necessary to make the flow path from the purge control valve 161 to the fuel tank 51 a closed space.

  In addition, a pressure sensor 63 (pressure detection means) is provided in the canister side second passage 56a, and the pressure sensor 63 outputs a voltage proportional to the pressure (absolute pressure) of the flow path that is closed in the leak diagnosis.

  In the engine controller 22 comprising a microcomputer, whether or not there is a leak in the flow path from the fuel tank 51 to the purge control valve 61 using the two valves (purge control valve 61 and drain cut valve 62) and the pressure sensor 63. This diagnosis is performed while the engine is running. That is, the drain cut valve 62 and the purge control valve 61 are fully closed, and the flow path from the fuel tank 51 to the purge control valve 61 is a closed space in which there is a relative pressure difference with respect to atmospheric pressure, and Since the presence or absence of a leak can be seen by looking at the pressure change after the closed space has a pressure difference relative to the atmospheric pressure, the intake pressure (intake pipe pressure downstream of the throttle 42 valve) is used. A pull-down process for reducing the flow path from the fuel tank 51 to the intake pipe 57 to a constant pressure; and a leak-down process for holding the flow path from the fuel tank 51 to the intake pipe 57 as a closed space after the pull-down process. In the leak down process, the pressure sensor 63 is used to sample the pressure in the flow path from the fuel tank 51 to the intake pipe 57. It is determined whether there is leakage based on the pressure sampled values.

  However, when the atmospheric pressure is greatly reduced due to traveling at high altitude, a diagnosis determination is made in determining whether there is a leak. For example, when there is a leak, the atmosphere enters the closed space in which the flow path from the fuel tank 51 to the purge control valve 61 is set to a pressure lower than the atmospheric pressure, so the flow path from the fuel tank 51 to the purge control valve 61 is It is expected that the pressure after a predetermined time after the closed space is larger than the pressure when the closed space is exceeded by a predetermined value. However, due to a significant drop in atmospheric pressure, the pressure in the closed space and the atmospheric pressure do not change much. Then, even if there is a leak, the atmosphere does not enter the closed space so much that the pressure after a predetermined time may not exceed the predetermined value beyond the pressure when the closed space is set. . In this case, it is erroneously determined that there is no leak even though there is a leak.

  For this reason, one of the leak diagnosis permission conditions is that the atmospheric pressure is equal to or higher than a predetermined value. When the atmospheric pressure is less than the predetermined value, the leak diagnosis is prohibited because the leak diagnosis permission condition is not satisfied. Here, the altitude when the atmospheric pressure becomes a predetermined value is about 2400 m.

  Thus, the atmospheric pressure is used for determining whether or not the leak diagnosis permission condition is satisfied.

  The leak diagnosis is included in the engine control, and the atmospheric pressure is not limited to the engine control. In this case, the engine is a fuel injection control using a returnless fuel supply line. It is also used for correcting the fuel injection pulse width (fuel injection amount).

  Explaining this, the fuel supply line to the engine 2 in FIG. 3 is a so-called returnless system without a line to return to the fuel tank 51. Since this returnless type fuel supply line is known from Japanese Patent Application Laid-Open No. 2001-107776, the fuel pump unit 71 is provided in the fuel tank 51 in brief. The fuel pump unit 71 is a unit in which a fuel pump, a fuel filter, and a fuel pressure regulator are integrated, and the fuel in the fuel tank 51 is supplied to the fuel injection valve 43 via the fuel pipe 72 by the fuel pump. The fuel pressure regulator functions as a control valve for maintaining the fuel supplied to the fuel injection valve 43 in a high pressure state of a specified pressure. When the fuel pressure of the fuel pipe 72 exceeds the specified pressure, the fuel pipe 72 The high-pressure fuel inside is returned to the fuel tank 51 through this fuel regulator. Thus, the fuel pressure regulator maintains the fuel pressure in the fuel pipe 72 at a constant pressure relative to the tank internal pressure.

  According to such a returnless type fuel supply line, the amount of fuel supplied per injection injected from the fuel injection valve 43 is equal to the fuel pressure if the valve opening pulse width of the fuel injection valve 43 is constant. It is proportional to the pressure difference between the intake pressure (intake pipe pressure downstream of the throttle valve 42), specifically, the atmospheric pressure as the pressure in the fuel tank 51 and the intake pressure. That is, even if the valve opening pulse width of the fuel injection valve 43 is constant, when the intake pressure is lower than the atmospheric pressure as in a low load, the intake pressure is close to the atmospheric pressure as in a high load. The amount of fuel supplied per injection increases, and the amount of fuel supplied per injection injected from the fuel injection valve 43 differs depending on the difference in intake pressure.

  Therefore, a correction coefficient KBST corresponding to the pressure difference between the atmospheric pressure and the intake pressure is introduced, and the fuel injection pulse width is corrected by the correction coefficient KBST as shown in the following equation.

Ti = (Tp × Tfbya + Kathos) × (α + αm−1)
× KBST × 2 + Ts (1)
Where Ti: fuel injection pulse width,
Tp: basic injection pulse width,
Tfbya: target equivalent ratio,
Kathos: Transient correction amount,
α: Air-fuel ratio feedback correction coefficient,
αm: air-fuel ratio learning value,
Ts: Invalid injection pulse width,
For example, the correction coefficient KBST has a characteristic as shown in FIG. 10, and becomes 1.0 when the intake pressure is equal to the atmospheric pressure. At this time, the fuel injection pulse width is not corrected. That is, the basic injection pulse width Tp is adapted when the intake pressure is equal to the atmospheric pressure.

  At low load when the intake pressure is smaller than atmospheric pressure and the amount of fuel supplied per injection is large, the correction coefficient KBST is a positive value smaller than 1.0, and the fuel injection pulse width is corrected to the decrease side. Further, when the intake pressure approaches high atmospheric pressure, the correction coefficient KBST is smaller than 1.0 but becomes a positive value close to 1.0, and the fuel injection pulse width is corrected to the decrease side. Is not corrected for weight loss. Thereby, if the valve opening pulse width of the fuel injection valve 43 is constant, the amount of fuel supplied per injection does not change regardless of the engine load state.

  Thus, when the engine is an engine that performs fuel injection control using a returnless fuel supply line, the atmospheric pressure is also used for correcting the fuel injection pulse width (fuel injection amount).

  In order to determine whether or not the leak diagnosis permission condition is satisfied, it is also necessary to measure the atmospheric pressure in order to perform fuel injection control using a returnless fuel supply line.

  In this case, since a new sensor for detecting atmospheric pressure increases the cost, in a vehicle using only the engine as a power source, the atmospheric pressure can be detected by the pressure sensor 63 used for leak diagnosis. (See JP 2003-35216 A). When the purge control valve 61 is fully closed during engine operation, the canister-side second passage 56a is opened to the atmosphere via the atmosphere opening 55, so that the purge control valve 61 is fully closed. Sometimes, the flow path pressure detected by the pressure sensor 63 is taken in as an atmospheric pressure measurement value.

  In this way, in a vehicle using only the engine as a power source, the atmospheric pressure can be measured by the pressure sensor 63. However, when this same engine that includes an evaporative fuel processing device and performs a leak diagnosis is used in a hybrid vehicle, the canister It has been newly found that there is a situation in which atmospheric pressure cannot be measured easily and accurately. That is, in order to measure the atmospheric pressure by the pressure sensor 63, it is necessary to capture the opportunity for the purgin control valve 61 to be fully closed. However, in an engine used in a hybrid vehicle, unlike a vehicle using only the engine as a power source, the engine is less likely to be operated while the vehicle is running. Therefore, in an engine used in a hybrid vehicle, During the operation, the purge control valve 61 can be said to be left open for the purpose of purging the gasoline particles from the canister 54, and there is no chance of the purge control valve 61 being fully closed. Even so, when the atmospheric pressure is to be measured, the purge control valve 61 is not fully closed, that is, the pressure at the low purge flow rate is taken in as the atmospheric pressure measurement value. However, the low purge flow rate (during the purge) means that if gasoline particles are adsorbed on the activated carbon 54a, it means that the flow resistance becomes a flow resistance and a pressure loss occurs in the canister 54. If the pressure monitored at the flow rate is the atmospheric pressure measurement value, an error corresponding to the pressure loss of the canister 54 occurs in the atmospheric pressure measurement value.

  On the other hand, in the hybrid vehicle, the engine 2 is automatically stopped (idle stop) when the idle stop permission condition is satisfied while the vehicle is running, and then the engine 2 is stopped when the idle stop permission condition is not satisfied. Since it has a function of automatically restarting, the purge control valve 61 is fully closed at an idle stop when the engine operation is automatically stopped. For this reason, the pressure monitored during idle stop (during purge stop) can be considered as the atmospheric pressure measurement value. However, when gasoline particles are adsorbed on the activated carbon 54a and pressure loss occurs in the canister 54, the idle stop is performed. Even if time has come, the pressure in the canister side second passage 56a does not immediately become atmospheric pressure, but the pressure in the canister side second passage 56a converges to atmospheric pressure with a response delay. Therefore, this response delay time is set as a delay time, and the pressure after the delay time has elapsed after the idling stop is taken in as the atmospheric pressure measurement value, so that an accurate atmospheric pressure measurement value can be obtained. It is possible.

  However, in this method, the more the gasoline particles adsorbed on the activated carbon 54a, the longer the response delay time until the pressure in the canister-side second passage 56a converges to atmospheric pressure. If a short delay time is set, an error occurs in the atmospheric pressure measurement value. However, if the delay time is increased, a situation in which the atmospheric pressure measurement value cannot be taken in occurs when the idle stop time is short.

  As described above, as long as the method of introducing the atmosphere to the flow path to the purge control valve 61 through the canister 54 is concerned, the hybrid vehicle with few opportunities to operate the engine is greatly affected by the pressure loss of the canister 54. Therefore, the atmospheric pressure cannot be measured easily and accurately.

  Therefore, in this embodiment, the point that the atmospheric pressure measurement value is taken in at the time of idling stop (when the engine is automatically stopped) is the same, but the purge control valve 61 is opened at idling stop, and the flow path pressure at this time is changed. Capture as atmospheric pressure measurements. Since there is no cause of pressure loss in the flow path from the purge control valve 61 to the air cleaner 74 during idle stop, the pressure in the flow path from the purge control valve 61 to the air cleaner 74 becomes atmospheric pressure during idle stop. Yes. Specifically, although the air cleaner 74 and the throttle valve 42 cause a slight pressure loss, the pressure loss is minute and at a level that is not a problem. Accordingly, when the purge control valve 61 is opened at the time of idling stop, the second passage 56 becomes the atmospheric pressure. At this time, if the flow path pressure monitored by the pressure sensor 63 is taken in as the atmospheric pressure measurement value, the atmospheric pressure is simple and has few errors. Can be obtained.

  This control executed by the engine controller 21 will be described in detail based on the following flowchart.

  FIG. 4 is for measuring the atmospheric pressure during engine rotation, and is executed at regular time intervals (for example, every 100 ms).

  In step 001, the flow path pressure P detected by the pressure sensor 63 is read.

  In step 002, the purge control valve opening (abbreviated as “purge valve opening” in the figure) and the threshold are compared.

  The purge control valve 61 is opened mainly in a region where the throttle valve downstream 42 is smaller than the atmospheric pressure (low load region). In this region, the purge control valve opening is determined according to the operating conditions.

  The threshold value is a value for checking whether or not the purge control valve 61 is in a fully closed state in a vehicle using only the engine as a power source. However, since the engine is applied to a hybrid vehicle, the threshold value is low. Use the upper limit of flow rate. Therefore, even if the purge control valve 61 is not fully closed, the atmospheric pressure measurement value during engine rotation can be taken in at a low purge flow rate. As described above, the reason why the conditions for capturing the atmospheric pressure measurement value during engine rotation are expanded to the time of the low purge flow rate is to increase the opportunity for capturing the atmospheric pressure measurement value during engine rotation. This low purge flow rate includes when the purge flow rate is zero, that is, when the purge control valve 61 is fully closed.

  The threshold value that determines the upper limit of the low purge flow rate is determined by adaptation. Even if the threshold value is determined, if the purge control valve opening never falls below the threshold value during vehicle operation, the measured atmospheric pressure value during engine rotation in step 007 cannot be obtained, and step 213 in FIG. It becomes impossible to determine the atmospheric pressure during engine rotation at the engine, that is, it becomes impossible to measure the atmospheric pressure during engine rotation. For this reason, it is necessary to set the threshold value so that the purge control valve opening is less than the threshold value at least once before the end of driving of the vehicle.

  However, if the measured atmospheric pressure value during engine rotation is taken at a low purge flow rate, an error corresponding to the pressure loss of the canister 54 will occur in the measured atmospheric pressure value during engine rotation at this time. This will be described later. This can be solved by introducing an atmospheric pressure correction amount learning value.

  When the purge control valve opening is less than the threshold, it is determined that the purge flow is at a low purge flow rate, and the process proceeds to step 003 to check whether the purge control valve opening was less than the previous threshold. In the previous time, when the purge control valve opening was equal to or greater than the threshold value, that is, when the purge control valve opening was less than the current threshold value, it is determined that the switching was performed at the low purge flow rate. Starts (atmospheric pressure measurement delay timer 1 value = 0). This atmospheric pressure measurement delay timer 1 is for measuring the elapsed time from the timing switched at the time of a low purge flow rate.

  If the purge control valve opening was previously less than the threshold value in step 003, that is, if the purge control valve opening was continuously less than the threshold value, the process proceeds from step 003 to step 005, and the atmospheric pressure measurement delay timer 1 value and delay time It is checked whether the atmospheric pressure measurement delay condition is satisfied by comparing 1. When the atmospheric pressure measurement delay timer 1 is activated last time and the process proceeds to step 005 this time, the atmospheric pressure measurement delay timer 1 value is less than the delay time 1 (that is, the atmospheric pressure measurement delay condition is not satisfied). Proceeding to 006, the atmospheric pressure measurement delay timer 1 value is incremented by the calculation period (100 ms).

  Thereafter, when the purge control valve opening is continuously less than the threshold value, the operation of step 006 is repeated, and eventually the atmospheric pressure measurement delay timer 1 value becomes the delay time 1 or more in step 005 (atmospheric pressure measurement delay condition) Meet). At this time, the routine proceeds from step 005 to step 007, where the flow passage pressure P is transferred to the measured atmospheric pressure value during engine rotation, and the measured atmospheric pressure value during engine rotation is stored in a predetermined memory.

  The reason for providing the delay time 1 is as follows. That is, before switching to the low purge flow rate, the intake pressure (pressure downstream of the throttle valve 42) is guided to the second passage 56. Therefore, immediately after switching to the low purge flow rate, the canister side purge passage is switched. The predetermined time is provided as the delay time 1 because the pressure of 56a is not stabilized and has a predetermined time until it is stabilized.

  This completes the measurement of the atmospheric pressure during the rotation of the engine. Therefore, in step 008, the atmospheric pressure measurement end flag 1 (initially set to zero when the engine is started) = 1 is set.

  On the other hand, when the purge control valve opening is equal to or greater than the threshold value in step 002, the process proceeds to step 009, where the atmospheric pressure measurement delay timer 1 is cleared (atmospheric pressure measurement delay timer 1 value = 0), and the atmospheric pressure measurement value during engine rotation is obtained. Stop importing.

  Thus, the flow path pressure after a predetermined delay time 1 has elapsed since switching at the low purge flow rate is stored as the measured value of atmospheric pressure during engine rotation.

  FIG. 5 is for generating a timer for measuring an elapsed time from the timing when the atmospheric pressure during engine rotation is measured, and is executed at regular intervals (for example, every 100 ms). The flow of FIG. 5 is executed following the flow of FIG.

  In step 111, the atmospheric pressure measurement end flag 1 set by the flow of FIG. When the atmospheric pressure measurement end flag 1 = 1, the routine proceeds to step 112, and it is checked whether or not the atmospheric pressure measurement end flag 1 = 1 was last time. When the atmospheric pressure measurement end flag 1 was previously set to 0, that is, when the atmospheric pressure measurement end flag 1 was switched to 1 at this time, the process proceeds to step 113, and an elapsed time timer after atmospheric pressure measurement is started (after atmospheric pressure measurement). Elapsed time timer value = 0). The elapsed time timer after atmospheric pressure measurement is for measuring the elapsed time from the timing at which the atmospheric pressure during engine rotation is measured.

  If the atmospheric pressure measurement end flag 1 = 1 in the previous step 112, that is, if the atmospheric pressure measurement end flag 1 = 1 continues, the routine proceeds from step 112 to step 114, where the elapsed time timer value after atmospheric pressure measurement is calculated as the calculation cycle. Increment by (100 ms).

  On the other hand, when the atmospheric pressure measurement end flag 1 = 0 in step 111, the routine proceeds to step 115, where the elapsed time timer after atmospheric pressure measurement is cleared (the elapsed time timer value after atmospheric pressure measurement = 0).

  Thus, the elapsed time from the timing at which the atmospheric pressure during engine rotation is measured is measured by the elapsed time timer value after atmospheric pressure measurement.

  FIG. 6 is for measuring the atmospheric pressure at the time of idling stop, and is executed at regular intervals (for example, every 100 ms).

  In step 101, the flow path pressure P detected by the pressure sensor 63 is read.

  In step 102, it is checked whether or not an idle stop permission condition is satisfied. The idle stop permission condition is, for example, that all the following conditions [1] to [5] are satisfied (see Japanese Patent Application Laid-Open No. 2004-076599).

  [1] The engine coolant temperature detected by the water temperature sensor 37 is within an appropriate range (for example, after completion of warm-up).

  [2] No charge request has been issued.

  [3] The master back (trademark) pressure detected by the intake pressure sensor 38 is not more than a predetermined value.

  [4] The steering angle detected by the steering angle sensor 39 is not large.

  [5] The transmission shift position detected by the shift position sensor 36 is not in the R range.

  Thus, the idle stop permission condition does not include the condition that the vehicle speed = 0 km / h and the brake is operating. That is, the engine 2 is automatically stopped mainly while the vehicle is traveling, and restart after the engine is automatically stopped is also performed while the vehicle is traveling.

  When all the conditions [1] to [5] are satisfied, it is determined that the idle stop permission condition is satisfied, and the process proceeds to step 103 to check whether or not the previous time was the idle stop permission condition. When it was not the idling stop permission condition last time, that is, when the idling stop permission condition was met for the first time this time, the routine proceeds to step 104, where a timer is started after the engine is stopped (time value after engine stop = 0). This timer after engine stop is for measuring the elapsed time from the timing when the engine is switched to the idle stop permission condition.

  In step 105, the atmospheric pressure correction amount learning value calculated flag = 0 is set. The atmospheric pressure correction amount learning value calculated flag is a flag that is necessary in the flow of FIG. The reason why the atmospheric pressure correction amount learned value calculation completed flag = 0 is set at the timing of switching to the idle stop permission condition is to update an atmospheric pressure correction amount learned value described later for each idle stop (see FIG. 7). .

  When the idling stop permission condition was also set at the previous time in step 103, that is, when the idling stop permission condition was continued, the process proceeds from step 103 to step 106, and the engine stop delay is compared by comparing the timer value after engine stop with the delay time 2. Check whether the conditions are met. The timer is started after the engine was stopped last time, and when the routine proceeds to step 106 this time, the timer value after the engine stop is less than the delay time 2 (that is, the delay condition after engine stop is not satisfied). After the engine stops, the timer value is incremented by the calculation cycle (100 ms).

  After that, when the timer value after engine stop is continuously less than the delay time 2, the operation of step 107 is repeated, and eventually the timer value after engine stop becomes the delay time 2 or more at step 106 (delay condition after engine stop). Meet). At this time, the routine proceeds from step 106 to step 108, where it is checked whether the delay condition after engine stop was satisfied last time. When the delay condition after the engine stop was not satisfied last time, that is, when the delay condition after the engine stop was satisfied this time, the process proceeds to Steps 109 and 110, and the purge control valve 61 is fully opened (or a predetermined purge control valve opening state). And the atmospheric pressure measurement delay timer 2 is started (atmospheric pressure measurement delay timer 2 value = 0). The atmospheric pressure measurement delay timer 2 is for measuring the elapsed time from the timing when the purge control valve 61 is switched to the fully open state.

  In this way, the purge control valve 61 is switched from the fully closed state to the fully opened state after the elapse of the delay time 2 after the idle stop permission condition is satisfied.

  The reason why the purge control valve 61 is not fully opened at the timing when the idle stop permission condition is satisfied, but waits for the delay time 2 is as follows. That is, the engine is operated before the idle stop permission condition is satisfied, and the intake pressure is guided to the second passage 56 during the engine operation. Accordingly, the idle stop permission condition is satisfied and the engine is automatically stopped, and immediately after the purge control valve 61 is fully closed, the canister-side second passage 56a is still in a pressure state lower than the atmospheric pressure. In this case, if a large pressure loss occurs in the canister 54 because a lot of gasoline particles are adsorbed on the activated carbon 54a or the activated carbon 54a is contaminated, the pressure in the canister side second passage 56a is delayed in response. Therefore, this response delay time is provided as the delay time 2. Since the response delay time depends on the state of the canister 54 (the amount of gasoline particles adsorbed on the activated carbon 54a and the degree of contamination of the activated carbon 54a), an optimal value is set as the delay time 2. As a result, the canister-side second passage 56a is in the atmospheric pressure state after the elapse of the delay time 2 after the idle stop permission condition is satisfied.

  If the timer value after the engine stop is equal to or longer than the delay time 2 in the previous step 108 (the delay condition after the engine stop is satisfied), the process proceeds from step 108 to step 111 to compare the atmospheric pressure measurement delay timer 2 value with the delay time 3 To see if the atmospheric pressure measurement delay condition is satisfied. When the atmospheric pressure measurement delay timer 2 is activated last time and the process proceeds to steps 108 and 111 this time, the atmospheric pressure measurement delay timer 2 value is less than the delay time 3 (that is, the atmospheric pressure measurement delay condition is not satisfied). In step 112, the atmospheric pressure measurement delay timer 2 value is incremented by the calculation period (100 ms).

  Thereafter, when the atmospheric pressure measurement delay timer 2 value is continuously less than the delay time 3, the operations of steps 111 and 112 are repeated, and eventually the atmospheric pressure measurement delay timer 2 value becomes the delay time 3 or more in step 111. (Atmospheric pressure measurement delay condition is met). At this time, the routine proceeds from step 111 to step 113, where the flow path pressure P is transferred to the measured atmospheric pressure value at idle stop, and the measured atmospheric pressure value at idle stop is stored in a predetermined memory.

  In this way, the flow path pressure after the delay time 3 has elapsed after the purge control valve 61 is fully opened is taken in as an atmospheric pressure measurement value at the time of idle stop.

  Instead of taking the flow path pressure at the timing when the purge control valve 61 is fully opened as the atmospheric pressure measurement value at idle stop, the flow path pressure after waiting for the delay time 3 is taken as the atmospheric pressure measurement value at idle stop. The reason is as follows. That is, even if the purge control valve 61 is fully opened at the timing when the pressure in the canister-side second passage 56a returns to atmospheric pressure after the engine is automatically stopped, the pressure in the flow path from the air cleaner 74 to the purge control valve 61 It is theoretically possible that the pressure does not return to atmospheric pressure immediately (for example, the air cleaner 74 is clogged). In this case, since the pressure in the flow path from the air cleaner 74 to the purge control valve 61 returns to atmospheric pressure with a response delay, this response delay time is provided as the delay time 3. An appropriate value is set as the delay time 3. As a result, after the delay time 3 has elapsed since the purge control valve 61 is fully opened, the second passage 56 is in the atmospheric pressure state, and the flow path pressure at this time is taken in as an atmospheric pressure measurement value at the time of idling stop. Yes.

  However, it is rare if a pressure loss that must be taken into consideration occurs in the flow path from the air cleaner 74 to the purge control valve 61, and if this rare case can be ignored, the delay time 3 = 0. That's fine.

  This completes the measurement of the atmospheric pressure at the time of idling stop. Therefore, in steps 114 and 115, the atmospheric pressure measurement end flag 2 (initially set to zero when the engine is started) is set to 1, and the purge control valve 61 is fully closed. Output a request to

  On the other hand, when the idle stop permission condition is not satisfied in step 102, the process proceeds to steps 116, 117, and 118, and the timer and the atmospheric pressure measurement delay timer 2 after the engine stop are cleared (timer value after engine stop = 0, atmospheric pressure measurement delay timer 2 value). = 0) and a request for fully closing the purge control valve 61 is output. In addition, even when the idle stop permission condition is no longer satisfied, the operation of steps 116 and 118 is executed from step 102 to stop taking in the atmospheric pressure measurement value at idle stop.

  As described above, when the idle permission condition is satisfied and the engine is automatically stopped, and when the delay time 2 elapses from the timing when the purge control valve 61 is fully closed, the purge control valve 61 is fully opened (or a predetermined purge is performed). The flow path pressure at the timing when the delay time 3 has elapsed after the purge control valve 61 is switched to the fully open state is stored in the memory as an atmospheric pressure measurement value at idle stop.

  FIG. 7 is for calculating the atmospheric pressure correction amount learning value, and is executed at regular intervals (for example, every 100 ms). The flow of FIG. 7 is executed following the flow of FIG.

  In step 121, the atmospheric pressure correction amount learned value calculation completed flag is viewed. This atmospheric pressure correction amount learned value calculated flag is a flag that is initially set to zero when the ignition switch is switched from OFF to ON during vehicle operation. Here, assuming that the atmospheric pressure correction amount learned value calculation flag = 0, the process proceeds to step 122 and subsequent steps.

  In steps 122 to 125, it is checked whether or not the following four conditions <1> to <4> are satisfied. When all of the following four conditions <1> to <4> are satisfied, it is determined that the atmospheric pressure correction amount learning value update condition (learning condition) is satisfied, and the process proceeds to step 126 and subsequent steps. If any one of the two conditions is not satisfied, it is determined that the update condition for the atmospheric pressure correction amount learning value is not satisfied, and the current process is terminated.

  <1> It is an idle stop permission condition (step 122).

  <2> The post-atmospheric pressure measurement elapsed time timer value generated in the flow of FIG. 5 is less than the learning permission time (step 123).

  <3> Atmospheric pressure measurement end flag 1 = 1 (step 124).

  <4> Atmospheric pressure measurement end flag 2 = 1 (step 125).

  The above conditions <3> and <4> are used to confirm that both the measured atmospheric pressure value during engine rotation and the measured atmospheric pressure value during idle stop are obtained at the present idle stop. The above <2> is used for the following reason. That is, even when both the atmospheric pressure measurement value during engine rotation and the atmospheric pressure measurement value during idle stop are obtained, the timing when the atmospheric pressure measurement value during engine rotation is obtained, and the timing when the atmospheric pressure measurement value during idle stop is obtained. If there is a large time difference between the two, the vehicle state and environmental conditions (altitude) may have changed significantly during that time. For example, atmospheric pressure is greatly affected by altitude, so even if the vehicle condition is the same, the vehicle will go up and down a steep slope after the timing when the measured atmospheric pressure is measured during engine rotation, and the atmospheric pressure during idle stop If the timing at which the measurement value is obtained has been reached, an error occurs in the atmospheric pressure measurement value at the time of idling stop by the difference in altitude between the two timings. Therefore, if a time that is not so long is set as the learning permission time, the vehicle state and the environmental condition (altitude) at the timing when the two atmospheric pressure measurement values are obtained can be regarded as substantially the same. In other words, the state shifts to idle stop immediately after the timing when the atmospheric pressure measurement value during engine rotation is obtained, and the condition <2> is satisfied when the atmospheric pressure measurement value during idle stop is obtained during the idling stop. It will go on later.

  In step 126, using the measured atmospheric pressure value during engine rotation obtained in step 007 of FIG. 4 and the measured atmospheric pressure value during idling stop obtained in step 113 of FIG. Is calculated.

Atmospheric pressure correction amount = Atmospheric pressure measurement value at idle stop-Atmospheric pressure measurement value during engine rotation
... (2)
As described above, since the upper limit value of the low purge flow rate is adopted as the threshold value of steps 002 and 003 in FIG. There is an error in loss. Therefore, the pressure loss of the canister 54 at that time is obtained as the atmospheric pressure correction amount by the equation (2). In addition, the vehicle state and environmental conditions (altitude) have not changed significantly between the timing when the atmospheric pressure measurement value during engine rotation is obtained and the timing when the atmospheric pressure measurement value during idle stop is obtained. The atmospheric pressure correction amount as the pressure loss of the canister 54 can be obtained with high accuracy from the equation (2).

  In Step 127, the weighted average value of the atmospheric pressure correction amount is used as the atmospheric pressure correction amount learning value, that is, the atmospheric pressure correction amount learning value is calculated by the following equation, and the calculated atmospheric pressure correction amount learning value ends the idle stop. Then, in step 128, the data is stored in a predetermined memory such as an EEPROM so as not to disappear even after the engine is automatically restarted.

Atmospheric pressure correction amount learning value = Atmospheric pressure correction amount learning value (previous value) × (1−learning coefficient)
+ Atmospheric pressure correction amount × learning coefficient (3)
However, the atmospheric pressure correction amount learning value (previous value): the previous value of the atmospheric pressure correction amount learning value,
Learning coefficient: constant between 0 and 1,
Here, an appropriate value is set as the initial value of the atmospheric pressure correction amount learning value (previous value).

  This completes the calculation of the atmospheric pressure correction amount learning value, so that the atmospheric pressure correction amount learning value calculated flag = 1 is set in step 129 to prepare for the next idle stop. Due to the atmospheric pressure correction amount learned value calculation completed flag = 1, it is not possible to proceed from step 121 to step 122 onward from the next time. That is, after turning on the ignition switch and starting driving the vehicle and obtaining the atmospheric pressure correction amount learning value at the subsequent idle stop, the atmospheric pressure correction amount learning value is held until the next idle stop. Keep it. Then, when the next idle stop is reached, the atmospheric pressure correction amount learned value calculation completed flag = 0 is set at step 105 in FIG. 6, so that it is possible to proceed from step 121 to step 122 onward in FIG. If all the conditions of 122 to 125 are satisfied, the atmospheric pressure correction amount learning value is updated in steps 126 and 127.

  As described above, the atmospheric pressure correction amount is calculated every time the idling stop is performed, and when the atmospheric pressure correction amount is calculated, the current pressure of the canister 54 is calculated by weighted averaging the atmospheric pressure correction amount learning value. An atmospheric pressure correction amount learning value corresponding to the loss can be obtained.

  FIG. 8 is for calculating the atmospheric pressure, and is executed at regular intervals (for example, every 100 ms) regardless of the engine rotation and idling stop.

  In step 211, it is determined whether or not the engine is rotating. When the engine is rotating, the routine proceeds to step 212 where the atmospheric pressure measurement end flag 1 is checked. If the atmospheric pressure measurement end flag 1 = 1, the measured value of atmospheric pressure during engine rotation has been obtained, so the process proceeds to step 213, and the measured atmospheric pressure value during engine rotation is stored in the memory at step 128 of FIG. The value obtained by adding the learned atmospheric pressure correction amount is set as the atmospheric pressure, that is, the atmospheric pressure is calculated by the following equation.

Atmospheric pressure = measured atmospheric pressure during engine rotation + learned atmospheric pressure correction value
... (4)
When the above is repeated, the upper limit value of the low purge flow rate is adopted as the threshold value of steps 002 and 003 in FIG. Has occurred. Accordingly, the pressure loss of the canister 54 at that time is obtained as the atmospheric pressure correction amount by the above equation (2). In addition, the vehicle state and environmental conditions (altitude) have not changed significantly between the timing when the atmospheric pressure measurement value during engine rotation is obtained and the timing when the atmospheric pressure measurement value during idle stop is obtained. The atmospheric pressure correction amount as the pressure loss of the canister 54 can be obtained with high accuracy by the above equation (2).

  Therefore, even if there is an error corresponding to the pressure loss of the canister 54 in the measured atmospheric pressure value during the engine rotation, the measured atmospheric pressure value during the engine rotation having an error corresponding to the pressure loss of the canister 54 as shown in the equation (4). By adding the atmospheric pressure correction amount learning value as the pressure loss of the canister 54, the atmospheric pressure can be obtained with high accuracy even during engine rotation.

  When the atmospheric pressure measurement end flag 1 = 0 in step 212, the atmospheric pressure measurement value during engine rotation has not been obtained yet, so the processing is ended as it is.

  When the engine is not rotating at step 211 (that is, at the time of idling stop), the routine proceeds to step 214 and the atmospheric pressure measurement end flag 2 is checked. If the atmospheric pressure measurement end flag 2 = 1, since the atmospheric pressure measurement value at the time of idling stop is obtained, the process proceeds to step 215, and the atmospheric pressure measurement value at the time of idling stop is set as the atmospheric pressure. During idling stop, the purge control valve 61 is opened to introduce atmospheric pressure into the canister-side second passage 56a via the intake pipe 57. At this time, the flow path pressure is measured as an atmospheric pressure measurement value (measurement of atmospheric pressure during idling stop). Value), it is not affected by the pressure loss of the canister 54. Therefore, the atmospheric pressure can be obtained with high accuracy even during idling stop.

  If the atmospheric pressure measurement end flag 2 = 0 in step 215, the atmospheric pressure measurement value at the time of idling stop has not been obtained yet, so the processing is ended as it is.

  In this way, atmospheric pressure can be obtained with high accuracy during engine rotation and during idle stop. The atmospheric pressure thus obtained is used for engine control (calculation of fuel injection pulse width and setting of a leak diagnosis permission flag) which will be described later.

  FIG. 9 is for calculating the fuel injection pulse width Ti, and is executed by the engine controller 22 at regular intervals (for example, every 100 ms) as a background job.

  In step 301, it is determined whether or not the engine is rotating. When the engine is rotating, the routine proceeds to step 302 where the intake pressure detected by the intake pressure sensor 38 and the atmospheric pressure calculated by the flow of FIG. 8 are read. The atmospheric pressure at this time is the atmospheric pressure obtained in step 213 in FIG.

  In step 303, a pressure difference is calculated from these intake pressure and atmospheric pressure by the following equation, and in step 304, a table having the contents shown in FIG. 10 is searched and a correction coefficient KBST is calculated.

Pressure difference = atmospheric pressure-intake pressure (5)
In step 305, using this correction coefficient KBST, for example, the fuel injection pulse width Ti at the time of sequential injection is calculated by the above equation (1).

  As shown in FIG. 10, since the intake pressure is smaller than the atmospheric pressure at low load, the correction coefficient KBST is a positive value smaller than 1.0. On the other hand, the intake pressure is higher at high load. Since the value is close to atmospheric pressure, the correction coefficient KBST is a positive value closer to 1.0 than when the load is low. That is, the basic injection pulse width Tp is corrected by decreasing the correction coefficient KBST, which is a positive value smaller than 1.0 at both low and high loads. However, the correction coefficient KBST is smaller at the low load. The reduction correction amount of the basic injection pulse width Tp is large.

  On the other hand, when the engine is not rotating (during idling stop), the process proceeds to step 306 to automatically stop the engine, and the invalid injection pulse width Ts is set to the fuel injection pulse width Ti. At this time, fuel is not injected from the fuel injection valve 43.

  Tp, Tfbya, Kathos, α, αm, and Ts in the above formula (1) are known values. To briefly explain the main values, the basic injection pulse width Tp is a value obtained by multiplying the value obtained by dividing the intake air amount Qa detected by the air flow meter 73 (see FIG. 3) by the engine speed by a constant K. Thus, an air-fuel mixture having a basic air-fuel ratio (substantially equal to the theoretical air-fuel ratio) is obtained by this basic injection pulse width Tp. Actually, manufacturing variations and deterioration with time occur in the flow characteristics of the air flow meter 73 and the flow characteristics of the fuel injection valve 43, and the air-fuel ratio deviates from the theoretical air-fuel ratio. The value for suppressing this variation is an air-fuel ratio feedback correction coefficient α calculated based on the output of the oxygen sensor 75 (see FIG. 3) for detecting the air-fuel ratio of the exhaust, and the air-fuel ratio of the exhaust is richer than the stoichiometric air-fuel ratio. If this is the case, the air-fuel ratio feedback correction coefficient α becomes smaller than 1.0, and the basic injection pulse width Tp is corrected to decrease, thereby returning the air-fuel ratio to the vicinity of the theoretical air-fuel ratio. On the contrary, if the air-fuel ratio of the exhaust gas is leaner than the stoichiometric air-fuel ratio, the air-fuel ratio feedback correction coefficient α is larger than 1.0, and the basic injection pulse width Tp is increased and corrected to increase the air-fuel ratio. Return the fuel ratio to near the stoichiometric air-fuel ratio.

  The target equivalent ratio Tfbya is a value required when the engine is operated at an air-fuel ratio leaner than the stoichiometric air-fuel ratio. For example, in the lean operation range, the target equivalent ratio Tfbya is a positive value smaller than 1.0. The basic injection pulse width Tp is corrected to decrease by this positive target equivalent ratio Tfbya smaller than 1.0. The injected fuel from the fuel injection valve 43 does not flow entirely into the combustion chamber by riding on the airflow, but part of it adheres to the intake port and the back of the intake valve umbrella to form wall flow fuel and burns slowly Flows into the chamber. Since the amount of this wall flow fuel has a large response delay at the time of transient, the air-fuel ratio is affected, so the value for correcting the supply delay to the combustion chamber at the time of this wall flow fuel transient is the transient correction amount. Kathos. The actual valve opening width of the fuel injection valve 43 is affected by the battery voltage. Even if the same basic injection pulse width Tp is instructed, the actual valve opening width decreases when the battery voltage decreases, and the desired air-fuel ratio is obtained. Therefore, the invalid injection pulse width Ts is a value for obtaining a desired air-fuel ratio even when the battery voltage is lowered.

  FIG. 11 is for setting a leak diagnosis permission flag, which is executed by the engine controller 22 at regular intervals (for example, every 100 ms).

  Steps 401 and 402 are the same as steps 301 and 302 in FIG. That is, in step 401, it is determined whether or not the engine is rotating. When the engine is rotating, the routine proceeds to step 402 where the basic injection pulse width Tp, the cooling water temperature Tw detected by the water temperature sensor 37, and the atmospheric pressure calculated by the flow of FIG. 8 are read. The atmospheric pressure at this time is the atmospheric pressure obtained in step 213 in FIG.

  In steps 403 to 405, it is checked whether or not the following three conditions <5> to <7> are satisfied. When all of the following three conditions <5> to <7> are satisfied, it is determined that the leak diagnosis permission condition is satisfied, and the routine proceeds to step 406 to set the leak diagnosis permission flag = 1. If any of these conditions is not satisfied, it is determined that the leak diagnosis permission condition is not satisfied, and the routine proceeds to step 407 where the leak diagnosis permission flag = 0 (leak diagnosis prohibited).

  <4> The basic injection pulse width Tp is equal to or less than the predetermined value 1 (step 403).

  <5> The coolant temperature Tw is equal to or higher than a predetermined value 2 (step 404).

  <6> The atmospheric pressure is not higher than the predetermined value 3, that is, not in a highland where the atmospheric pressure exceeds approximately 2400 m (step 405).

  In <4>, the basic injection pulse width Tp is equal to or less than the predetermined value 1, that is, when the load is low (the intake pressure is lower than the atmospheric pressure). This is because in order to guide the gasoline particles to the intake pipe 57, the second passage 56 needs to be in a pressure state lower than the atmospheric pressure. The reason why the coolant temperature Tw is less than the predetermined value 2 in <5>, that is, not in the state after completion of engine warm-up, is as follows. That is, if the combustion is originally unstable when the engine is not warmed up, and if fresh air (purge gas) mixed with gasoline particles is introduced, the combustion is further deteriorated. The reason why the atmospheric pressure is less than the predetermined value 3 in <6> is to avoid an erroneous determination in determining whether or not a leak has occurred.

  The leak diagnosis permission flag set in this way is used in a leak diagnosis routine (not shown). When the leak diagnosis permission flag = 1, the drain cut valve 62 and the purge control valve 61 are closed, and the flow path from the fuel tank 51 to the purge control valve 61 has a relative pressure difference with respect to the atmospheric pressure. A diagnosis is made as to whether or not there is a leak based on the pressure change after the closed space is made and the closed space has a pressure difference relative to the atmospheric pressure. In the middle of such leak diagnosis, when the vehicle moves to a high altitude exceeding 2400 m, the leak diagnosis permission flag is set to 0, so that the leak diagnosis is prohibited even during the leak diagnosis.

  Here, the effect of this embodiment is demonstrated.

  The canister 54 that causes pressure loss exists in the flow path from the purge control valve 61 to the atmosphere opening 55 of the canister 54, whereas the flow path from the purge control valve 61 to the air cleaner 74 upstream of the intake pipe 57. Since there is no cause of pressure loss, the pressure in the flow path from the purge control valve 61 to the air cleaner 74 upstream of the intake pipe 57 is at atmospheric pressure during idle stop (when the engine is automatically stopped). . According to the present embodiment (the invention described in claims 1 and 9), the purge control valve 61 is opened at the time of idling stop (when the engine is automatically stopped) to the air cleaner 74 upstream of the intake pipe 57 from the purge control valve 61. The pressure sensor 63 (pressure detection means) immediately comes into contact with the atmospheric pressure because the atmospheric pressure existing in the flow path is guided to the purge passage 56 (see step 109 in FIG. 6). Can be sought. At each idle stop, it is possible to obtain atmospheric pressure measurement values with little error. In other words, the second passage 56a communicates not only with the atmosphere via the atmosphere opening 55 of the canister 54 but also with the atmosphere via the intake pipe 57. By introducing the atmospheric air, it is possible to easily obtain the atmospheric pressure with few errors regardless of the pressure loss of the canister 54.

  According to the present embodiment (the invention described in claims 3 and 11), the flow path pressure at the low purge flow rate during engine rotation is stored as an atmospheric pressure measurement value during engine rotation (step 002 in FIG. 4). 003, 005, and 007), even when an engine that includes an evaporative fuel processing device and performs a leak diagnosis is used in a hybrid vehicle, the opportunity for taking in an atmospheric pressure measurement value during engine rotation increases.

  However, being at a low purge flow rate (during purge) means that pressure loss occurs in the canister 54 if gasoline particles are adsorbed on the activated carbon 54a. If the atmospheric pressure measurement value is used as it is, an error corresponding to the pressure loss of the canister 54 occurs in the atmospheric pressure measurement value. According to the present embodiment (the invention described in claims 3 and 11), At idle stop (when the engine is automatically stopped), the pressure difference between the measured atmospheric pressure value during idle stop (the measured atmospheric pressure value during automatic engine stop) and the measured atmospheric pressure during engine rotation is calculated as the atmospheric pressure correction amount. (Refer to steps 122 and 126 in FIG. 7) The atmospheric pressure correction amount represents the pressure loss of the canister 54. Accordingly, by calculating the value obtained by adding the calculated atmospheric pressure correction amount to the measured atmospheric pressure value during engine rotation as the atmospheric pressure during engine rotation (see steps 211, 212, and 213 in FIG. 8), the fuel vapor processing apparatus is When an engine that performs leak diagnosis is used in a hybrid vehicle, the atmospheric pressure can be accurately determined even during engine rotation as long as there is an opportunity to capture atmospheric pressure measurement values at low purge flow rates during engine rotation. it can.

  According to the present embodiment (the invention described in claims 4 and 12), the atmospheric pressure correction amount learning value is calculated based on the atmospheric pressure correction amount at every idle stop (when the engine is automatically stopped) (FIG. 7). Since the value obtained by adding the learned atmospheric pressure correction value to the measured atmospheric pressure value during engine rotation is calculated as the atmospheric pressure during engine rotation (see steps 211, 212, and 213 in FIG. 8), the canister Also, when the purge of gasoline particles from 54 progresses and the pressure loss of the canister 54 changes to a smaller side, the changed pressure loss is taken every idle stop (every time the engine is automatically stopped). Can be reflected in the atmospheric pressure correction amount learning value, and even if the pressure loss of the canister 54 changes to a smaller side during engine operation, the atmospheric pressure is accurate. It is possible to obtain.

  If the engine is an engine that performs fuel injection control using a returnless fuel supply line, and there is a measurement error in the atmospheric pressure measurement value during engine rotation, the error corresponding to the measurement error in the atmospheric pressure measurement value Is generated in the fuel injection pulse width Ti (fuel injection amount). According to the present embodiment (the invention described in claims 7 and 15), the atmospheric pressure during engine rotation is corrected to the fuel injection pulse width. 9 (see steps 301, 302, 303, 304, and 305 in FIG. 9), even in the case of an engine that performs fuel injection control using a return-less fuel supply line, the measurement error of the atmospheric pressure measurement value is reduced. The accompanying fuel injection amount error can be prevented.

  According to the present embodiment (the invention described in claims 8 and 16), the atmospheric pressure during engine rotation is used for determining whether or not the leak diagnosis permission condition is satisfied (steps 401, 405, and 406 in FIG. 11). 407), even in the case where the atmospheric pressure is measured by the pressure sensor 63 used for the leak diagnosis in the engine having the evaporative fuel processing device and performing the leak diagnosis, it is erroneously determined whether or not the leak diagnosis permission condition is satisfied. Can be avoided.

  The embodiment has been described in the case of a hybrid vehicle that drives a vehicle using at least one of a motor and an engine. However, the embodiment is a vehicle that uses only an engine as a drive source, and the engine is operated when a predetermined operating condition is satisfied. The present invention is also applicable to a vehicle having a function of automatically stopping and restarting the engine automatically when another predetermined driving condition is satisfied (inventions according to claims 1 and 9).

  In the embodiment, an atmospheric pressure correction amount learning value is calculated based on the atmospheric pressure correction amount every time the engine is automatically stopped (at every idle stop), and this atmospheric pressure correction amount learning value is used as the measured atmospheric pressure value during engine rotation. As explained in the case of calculating the added value as the atmospheric pressure during engine rotation, without introducing the atmospheric pressure correction amount learning value, the measured value of atmospheric pressure during engine automatic stop and the value during engine rotation The pressure difference from the atmospheric pressure measurement value is calculated as the atmospheric pressure correction amount, and the value obtained by adding the calculated atmospheric pressure correction amount to the atmospheric pressure measurement value during engine rotation is calculated as the atmospheric pressure during engine rotation. It does not matter (the invention according to claims 3 and 11).

  In the embodiment, during the engine rotation, the atmospheric pressure during the engine rotation is set as the atmospheric pressure, and during the idle stop, the measured atmospheric pressure value during the idle stop is set as the atmospheric pressure. When it can be considered that the environmental conditions (altitude) are almost the same, the measured atmospheric pressure value at the time of idling stop can be used as the atmospheric pressure during the engine rotation (the invention according to claims 5, 6, 13, 14). . Alternatively, a value obtained by adding the atmospheric pressure correction amount learning value (or the atmospheric pressure correction amount) to the measured atmospheric pressure value during engine rotation can be used as the atmospheric pressure at the time of idling stop. It is also conceivable to use the measured atmospheric pressure at idle stop with priority.

  The leak diagnosis processing procedure according to claim 1 is performed by the engine controller 22, the valve opening processing procedure is performed by steps 102 and 109 in FIG. 6, and the atmospheric pressure measurement value capturing processing procedure at the time of automatic engine stop is performed by steps 102 and 113 in FIG. 6. Each is fulfilled.

  The function of the leak diagnosis means according to claim 7 is the engine controller 22, the function of the valve opening means is the steps 102 and 109 of FIG. 6, and the function of the atmospheric pressure measurement value capturing means at the time of automatic engine stop is the step of FIG. 6. 102 and 113, respectively.

The schematic block diagram of the control apparatus of the vehicle of 1st Embodiment of this invention. The schematic block diagram of the control system of the vehicle of 1st Embodiment of this invention. The schematic block diagram of the control system of the evaporation fuel processing apparatus with which an engine is equipped. The flowchart for demonstrating the measurement of the atmospheric pressure during engine rotation. The flowchart for demonstrating the production | generation of the elapsed time timer after an atmospheric pressure measurement during engine rotation. The flowchart for demonstrating the measurement of atmospheric pressure at the time of idle stop. The flowchart for demonstrating the calculation of an atmospheric pressure correction amount learning value. The flowchart for demonstrating the calculation of atmospheric pressure. The flowchart for demonstrating the calculation of a fuel-injection pulse width. FIG. The flowchart for demonstrating the setting of a leak diagnosis permission flag.

Explanation of symbols

2 Engine 22 Engine controller 51 Fuel tank 54 Canister 56 Second passage (purge passage)
57 Intake pipe 61 Purge control valve 62 Drain cut valve 63 Pressure sensor (pressure detection means)

Claims (16)

In a vehicle having a function of automatically stopping the engine when a predetermined driving condition is satisfied, and automatically restarting the engine when another predetermined driving condition is satisfied,
The engine is
A canister for guiding and adsorbing fuel vapor generated in the fuel tank;
A purge passage communicating the canister and an intake pipe downstream of the throttle valve;
A purge control valve for opening and closing the purge passage;
A drain cut valve for opening and closing the atmosphere opening of the canister;
Pressure detecting means for detecting the pressure in the flow path from the fuel tank to the purge control valve,
A leak diagnosis processing procedure for diagnosing whether or not there is a leak using these purge control valve, drain cut valve and pressure detection means when the leak diagnosis permission condition is satisfied;
A valve opening procedure for opening the purge control valve when the engine is automatically stopped;
When the purge control valve is opened by this valve opening procedure, the pressure of the flow path detected by the pressure detection means is taken as the atmospheric pressure measurement value at the time of engine automatic stop, The vehicle control method characterized by including.   The vehicle control method according to claim 1, further comprising a motor, wherein the vehicle is driven using at least one of the motor and the engine. An engine rotation atmospheric pressure measurement value storage processing procedure for storing the pressure of the flow path detected by the pressure detection means at a low purge flow rate during engine rotation as an engine rotation atmospheric pressure measurement value;
An atmospheric pressure correction amount calculation processing procedure for calculating a pressure difference between the atmospheric pressure measurement value during the automatic engine stop and the atmospheric pressure measurement value during the engine rotation as the atmospheric pressure correction amount when the engine is automatically stopped;
3. An engine rotation atmospheric pressure calculation processing procedure for calculating a value obtained by adding the calculated atmospheric pressure correction amount to the engine rotation atmospheric pressure measurement value as an atmospheric pressure during engine rotation. The vehicle control method described. An atmospheric pressure correction amount learning value calculation processing procedure for calculating an atmospheric pressure correction amount learning value based on the atmospheric pressure correction amount at each automatic stop of the engine;
4. An engine rotation atmospheric pressure calculation process procedure for calculating a value obtained by adding the atmospheric pressure correction amount learning value to the engine rotation atmospheric pressure measurement value as the engine rotation atmospheric pressure. The vehicle control method described.   2. The method according to claim 1, wherein when the engine is an engine that performs fuel injection control using a returnless fuel supply line, the measured atmospheric pressure value at the time of automatic engine stop is used for correction of the fuel injection amount. The vehicle control method according to 2.   The vehicle control method according to claim 1 or 2, wherein the measured value of atmospheric pressure at the time of automatic engine stop is used to determine whether or not the leak diagnosis permission condition is satisfied.   5. The method according to claim 3, wherein when the engine is an engine that performs fuel injection control using a returnless fuel supply line, the atmospheric pressure during the rotation of the engine is used for correcting the fuel injection amount. The vehicle control method described.   The vehicle control method according to claim 3 or 4, wherein the atmospheric pressure during the engine rotation is used to determine whether or not the leak diagnosis permission condition is satisfied. In a vehicle having a function of automatically stopping the engine when a predetermined driving condition is satisfied, and automatically restarting the engine when another predetermined driving condition is satisfied,
The engine is
A canister for guiding and adsorbing fuel vapor generated in the fuel tank;
A purge passage communicating the canister and an intake pipe downstream of the throttle valve;
A purge control valve for opening and closing the purge passage;
A drain cut valve for opening and closing the atmosphere opening of the canister;
Pressure detecting means for detecting the pressure in the flow path from the fuel tank to the purge control valve,
A leak diagnosis means for diagnosing whether or not there is a leak using the purge control valve, the drain cut valve and the pressure detection means when the leak diagnosis permission condition is satisfied;
A valve opening means for opening the purge control valve when the engine is automatically stopped;
An engine automatic stop atmospheric pressure measurement value taking-in means for taking in the pressure of the flow path detected by the pressure detection means as the atmospheric pressure measurement value during engine automatic stop when the purge control valve is opened by the valve opening means; A control apparatus for a vehicle.   The vehicle control device according to claim 9, comprising a motor, and driving the vehicle using at least one of the motor and the engine. An engine rotation atmospheric pressure measurement value storage means for storing the pressure of the flow path detected by the pressure detection means at a low purge flow rate during engine rotation as an engine rotation atmospheric pressure measurement value;
An atmospheric pressure correction amount calculating means for calculating a pressure difference between the atmospheric pressure measurement value at the time of automatic engine stop and the atmospheric pressure measurement value during rotation of the engine as an atmospheric pressure correction amount at the time of the automatic stop of the engine;
11. An engine rotation atmospheric pressure calculation means for calculating a value obtained by adding the calculated atmospheric pressure correction amount to the engine rotation atmospheric pressure measurement value as an atmospheric pressure during engine rotation. Vehicle control device. Atmospheric pressure correction amount learning value calculation means for calculating an atmospheric pressure correction amount learning value based on the atmospheric pressure correction amount at each automatic stop of the engine;
12. The engine rotation atmospheric pressure calculation means for calculating a value obtained by adding the atmospheric pressure correction amount learning value to the engine rotation atmospheric pressure measurement value as an atmospheric pressure during engine rotation. Vehicle control device.   10. The method according to claim 9, wherein when the engine is an engine that performs fuel injection control using a returnless fuel supply line, the atmospheric pressure measurement value at the time of automatic engine stop is used for correcting the fuel injection amount. The vehicle control device according to 10.   The vehicle control device according to claim 9 or 10, wherein the measured atmospheric pressure value at the time of automatic engine stop is used for determining whether or not the leak diagnosis permission condition is satisfied.   The engine according to claim 11 or 12, wherein when the engine is an engine that performs fuel injection control using a returnless fuel supply line, the atmospheric pressure during the rotation of the engine is used for correcting the fuel injection amount. The vehicle control device described.   The vehicle control device according to claim 11 or 12, wherein the atmospheric pressure during the engine rotation is used for determining whether or not the leak diagnosis permission condition is satisfied.

Images