JP6006189B2 - Engine fuel injection control device - Google Patents

Description

  The present invention relates to an engine fuel injection control apparatus that performs O2 feedback control.

  2. Description of the Related Art Conventionally, an engine fuel injection control device that performs O2 feedback control using an output of an oxygen sensor that detects an oxygen concentration of exhaust gas is known (Patent Document 1). Specifically, the learning value obtained by the O2 feedback control is updated for each learning region defined by the engine speed and the throttle opening and divided according to the throttle opening. Then, the fuel injection amount is calculated using the learning value as the fuel injection correction coefficient in the operation in the learning region.

JP 2011-157830 A

  Here, when it is desired to perform lean operation with a thin air-fuel ratio in a conventional engine fuel injection control device in order to reduce fuel consumption, the following control can be considered. First, learning of the O2 feedback correction coefficient in the region where the lean operation is performed (lean operation region) is completed, and when the lean operation can be performed, the fuel injection amount is calculated using the learned value after learning. decide.

  Here, since the oxygen sensor is a sensor that detects whether the air-fuel ratio is lean or rich in the vicinity of the ideal air-fuel ratio, if the leaning operation is performed, the accurate air-fuel ratio cannot be determined. Therefore, O2 feedback control cannot be performed during lean operation. That is, it is difficult to detect even if the air-fuel ratio deviates from the assumed value during the lean operation.

  Whether or not the deviation of the air-fuel ratio during the lean operation occurs largely depends on the accuracy of the learned value in the lean operation region. Accordingly, a highly accurate learning value used during the lean operation is required.

  In a conventional fuel injection control device for an engine, a learning value is held for each learning region, and this learning region is set relatively wide. By setting a wide learning area, the speed to completion of learning can be improved. Note that, in each learning region, the learning value is often different between the low rotation speed end and the high rotation speed end, but the learning value is continuously updated during O2 feedback, so that the air-fuel ratio is sufficient. Accuracy can be ensured.

  However, since the learning value is not updated during leaning operation, the learning value learned before leaning operation is learned for leaning operation, depending on the operating conditions (throttle opening and engine speed) during leaning operation. In some cases, it was inappropriate to use as a value.

  Therefore, an object of the present invention is to provide an engine fuel injection control device that improves the accuracy of a learned value under a predetermined operating condition while maintaining the speed of completion of learning during O2 feedback control.

  The fuel injection control device (12) of the engine (10) according to the present invention has the following features.

  First feature: an oxygen sensor (50) for detecting an oxygen concentration in exhaust gas of the engine (10), and a throttle opening of a throttle valve (34) provided in an intake pipe (22) of the engine (10) When (TH) and the engine speed (NE) are within the O2 feedback region, a feedback correction amount (KO2) that is a correction coefficient is calculated based on the oxygen concentration detected by the oxygen sensor (50), and A first learning value (KBUave), which is a correction coefficient learned in accordance with the feedback correction amount (KO2), is expressed by the throttle opening (TH) and the engine speed (NE), and the throttle opening. It is calculated and updated for each of the plurality of first learning regions (A) of the first learning value map (54) in which the O2 feedback region is divided according to the degree (TH). When the number calculation unit (84), the throttle opening (TH), and the engine speed (NE) are within the O2 feedback region, the basic injection amount (TIMB) is set to the feedback correction amount (KO2) and In a fuel injection control device (12) of an engine (10), comprising: a fuel injection amount calculation unit (70) that calculates a fuel injection amount (TOUT) by correcting using the first learning value (KBUave). A plurality of second learning areas (in the first learning value map (54)) and at least the width of the engine speed (NE) is set narrower than that in the first learning value map (54). B), a second learning value map (56), and the coefficient calculation unit (84) is configured such that the throttle opening (TH) and the engine speed (NE) are in the second learning value map. 56), the second learning in the second learning region (B) of the second learning value map (56) corresponding to the throttle opening (TH) and the engine speed (NE) is further performed. The value (KBULEAN) is calculated and updated in accordance with the feedback correction amount (KO2), and the fuel injection amount calculation unit (70) determines that the throttle opening (TH) and the engine speed (NE) are the first values. 2 When performing a predetermined operation within the region of the learning value map (56), the basic injection amount (TIMB) is calculated using the second learning value (KBULEAN) instead of the first learning value (KBUave). The fuel injection amount (TOUT) is calculated after correction.

  Second feature: The predetermined operation is a lean operation, and the coefficient calculation unit (84) updates the second learning value (KBULEAN) when entering a lean mode in which the lean operation is possible. Stop.

  Third feature: During the lean mode, the lean operation is executed when the throttle opening (TH) and the engine speed (NE) are within the range of the second learning value map (56). The coefficient calculation unit (84) updates the first learning value (KBUave) when the leaning operation is not executed, and the first learning value (KBUave) when the leaning operation is executed. ) Stop updating.

  Fourth feature: In the second learning region (B2) where the throttle opening (TH) of the second learning value map (56) is on the low opening side, the larger the throttle opening (TH), The engine speed (NE) is set so as to widen.

  Fifth feature; the second learning region (B5) where the throttle opening (TH) of the second learning value map (56) is on the high opening side corresponds to the second learning region (B5). The upper limit of the throttle opening (TH) is set lower than the first learning area (A5) of the one learning value map (54).

  According to the first feature of the present invention, during the O2 feedback control, the first learning value of each first learning region in the first learning value map is updated to calculate the fuel injection amount, which is equivalent to the conventional case. The learning speed of the first learning value and the accuracy of the O2 feedback control can be ensured. In addition, when the second learning value map used for the predetermined operation is set in the region of the first learning value map and learning is performed in the second learning value map during the O2 feedback control, By learning both the first learning value of the first learning region and the second learning value of the second learning region in the second learning value map, even with the same throttle opening and engine speed, Different learning values can be learned. Therefore, the accuracy of the second learning value used for the predetermined operation is improved, and when performing the predetermined operation within the region of the second learning value map, the second learning value of the second learning value map is used to increase the accuracy. Accurate fuel injection can be performed.

  According to the second feature of the present invention, when entering the lean mode, the updating of the second learning value of each second learning region of the second learning value map before entering the lean mode is stopped. Deviation from the air-fuel ratio assumed during the lean operation can be suppressed.

  According to the third feature of the present invention, since the first learning value is updated when the leaning operation is not executed during the leaning mode, the accuracy of the first learning value is improved even during the leaning mode. Can be secured.

  According to the fourth feature of the present invention, the second learning area where the throttle opening of the second learning value map is on the low opening side is set so that the range of the engine speed increases as the throttle opening increases. As a result, the range in which lean operation can be performed is expanded as much as possible, and fuel consumption can be reduced.

  According to the fifth feature of the present invention, the second learning area where the throttle opening of the second learning value map is on the high opening side is the first learning area of the first learning value map corresponding to the second learning area. Since the upper limit of the throttle opening is set lower than that, the region in which the lean operation can be performed can be appropriately restricted, and the engine fluctuation during the lean operation can be suppressed.

It is a block diagram which shows the structure of the fuel-injection control apparatus of the engine of embodiment. It is a map for searching an engine operating region. It is a map which shows the feedback area | region of an air fuel ratio. FIG. 4 shows a first learning value map showing each first learning region of the O2 feedback region divided into six by overlapping regions defined in FIG. 2 and FIG. 3. It is a schematic diagram of the 1st learning value map shown in FIG. It is a schematic diagram of the 2nd learning value map which is a learning value map which shows each 2nd learning area | region of an open loop area | region. It is a block diagram which shows the structure of the control part of FIG. It is a figure which shows an example of the injection quantity map of FIG. It is a block diagram which shows the structure of the volatile memory of FIG. It is a flowchart which shows the setting operation | movement of the leaning mode in which leaning driving | operation is possible. 6 is a flowchart showing the operation of the fuel injection control apparatus when the throttle opening TH and the engine speed NE are within an O2 feedback region. It is a flowchart which shows operation | movement of update of KBUave etc. of a coefficient calculation part.

  DESCRIPTION OF EMBODIMENTS A fuel injection control device for an engine according to the present invention will be described in detail below with reference to the accompanying drawings and preferred embodiments.

  FIG. 1 is a block diagram illustrating a configuration of a fuel injection control device 12 of an engine 10 according to an embodiment. A piston 16 is slidably fitted to a cylinder bore 14 of the engine 10 mounted on a vehicle such as a motorcycle. An intake pipe 22 that supplies an air-fuel mixture to the combustion chamber 20 and an exhaust pipe 24 that discharges exhaust gas from the combustion chamber 20 are connected to the cylinder head 18 of the engine 10. A catalytic converter 26 is attached to the exhaust pipe 24. A spark plug 28, an intake valve 30, and an exhaust valve 32 whose front ends protrude into the combustion chamber 20 are attached to the cylinder head 18.

  The intake pipe 22 is provided with a throttle valve 34 that controls the amount of intake air so that it can be opened and closed, and a bypass passage 36 that connects the upstream side and the downstream side of the throttle valve 34. The bypass passage 36 is provided with an air screw 38 that adjusts the amount of air passing through the bypass passage 36 by opening and closing the bypass passage 36. The driver adjusts the air screw 38 to adjust the idling (idle) rotation speed. A fuel injection valve 40 that injects fuel is provided between the downstream side of the throttle valve 34 and the intake valve 30.

  The control unit 42 controls the ignition timing of the spark plug 28, the fuel injection amount injected from the fuel injection valve 40, and the like. The fuel injection control device 12 of the engine 10 further includes a throttle opening sensor 44 that detects the opening (hereinafter referred to as throttle opening) TH of the throttle valve 34 and the rotation of the crankshaft 10 a of the engine 10 connected to the piston 16. An engine speed sensor 46 for detecting the number (hereinafter referred to as engine speed) NE, a water temperature sensor 48 for detecting the coolant temperature TW of the engine 10, and oxygen remaining in the exhaust gas provided upstream of the catalytic converter 26. An oxygen sensor 50 for detecting the concentration and an intake air temperature sensor 52 for detecting the intake air temperature TA taken into the combustion chamber 20 are provided. Output signals of the throttle opening sensor 44, the engine speed sensor 46, the water temperature sensor 48, the oxygen sensor 50, and the intake air temperature sensor 52 are input to the control unit 42.

  FIG. 2 is a map for searching for an operation region of the engine 10. Based on the engine speed NE and the throttle opening TH, the control unit 42 searches for which region the operation region is in. In this map, the lower limit throttle opening THO2L and the upper limit throttle opening THO2H, and the plurality of throttle openings THFB0, THFB1, THFB2, and THFB3 between the two throttle openings increase as the engine speed NE increases. . These throttle openings are set in advance so that the relationship THO2L <THFB0 <THF1 <THB2 <THB3 <THO2H is satisfied.

  The set throttle openings THO2L, THFB0, THFB1, THFB2, THFB3, and THO2H are set with hysteresis. That is, the solid lines shown in FIG. 2 are boundary values applied when the throttle opening TH is increased, and the broken lines adjacent to the solid lines are boundary values applied when the throttle opening is decreased. is there.

  FIG. 3 is a map showing an air-fuel ratio feedback region. The air-fuel ratio feedback (O2 feedback, O2F / B) region indicated by hatching is represented by the engine speed NE and the throttle opening TH. The O2 feedback region is a region determined by a lower limit rotational speed NLOP, an idle region upper limit rotational speed NTHO2L, and an upper limit rotational speed NHOP, and a lower limit throttle opening THO2L and an upper limit throttle opening THO2H. The lower limit rotational speed NLOP, the idle region upper limit rotational speed NTHO2L, and the upper limit rotational speed NHOP are indicated by a solid line on the increasing side of the engine rotational speed NE and indicated by a broken line on the decreasing side of the engine rotational speed NE. Hysteresis is set. Hysteresis is set for the lower limit throttle opening THO2L and the upper limit throttle opening THO2H, in which the value on the increase side of the throttle opening TH is indicated by a solid line and the value on the decrease side of the throttle opening TH is indicated by a broken line. ing.

  FIG. 4 shows a first learning value map 54 showing the first learning regions A of the O2 feedback region divided into six by overlapping the regions defined in FIGS. 2 and 3. FIG. 5 is a schematic diagram of the first learning value map 54 shown in FIG. The first learning value map 54 is used for O2 feedback control.

  The plurality of first learning regions (feedback regions) A are obtained by dividing the O2 feedback region into six according to the operation region of FIG. 2, and are represented by the engine speed NE and the throttle opening TH. The plurality of first learning areas A are divided into six according to the throttle opening TH, and a plurality of operation areas including the plurality of first learning areas A based on the engine speed NE and the throttle opening TH. Is set. In the present embodiment, the operation regions of the first learning region A for the six O2 feedback controls are indicated with numbers “1” to “6”, and the operation regions other than the O2 feedback region are “0”. It is shown with numbers “7” to “11”.

  A first learning area A of “1” (hereinafter, first learning area A1) is an idling area of the engine 10. The first learning area A of “2” (hereinafter referred to as the first learning area A2) is the first learning area A that is adjacent to the first learning area A1 at the high engine speed NE. The first learning area A of “3” (hereinafter referred to as the first learning area A3) is a first learning area A adjacent to the first learning area A1 and the first learning area A2 on the high opening side of the throttle opening TH. is there. The first learning area A of “4” (hereinafter referred to as the first learning area A4) is the first learning area A adjacent to the first learning area A3 on the high opening side of the throttle opening TH. The first learning area A of “5” (hereinafter, first learning area A5) is the first learning area A adjacent to the first learning area A4 on the high opening side of the throttle opening TH. The first learning area A of “6” (hereinafter referred to as the first learning area A6) is the first learning area A adjacent to the first learning area A5 on the high opening side of the throttle opening TH.

  FIG. 6 is a schematic diagram of a second learning value map 56 that is a learning value map showing each second learning region B of the open loop region. The second learning value map 56 is used during a predetermined operation (lean operation in the present embodiment) that performs open loop control in which O2 feedback control cannot be performed. Lean operation refers to an operation (an operation with low fuel consumption) in which the air-fuel ratio is set higher than the ideal air-fuel ratio (14.7) in order to suppress fuel consumption (for example, the air-fuel ratio is 16 to 18). In FIG. 6, for easy understanding, the second learning value map 56 is displayed superimposed on the first learning value map 54 of a two-dot chain line. When the throttle opening TH and the engine speed NE enter an open loop region (hereinafter referred to as a lean region), in principle, a lean operation is performed by executing an open loop control (hereinafter referred to as a lean control).

  The plurality of second learning regions B are obtained by dividing the lean region into four, and are represented by the engine speed NE and the throttle opening TH. The lean region is set in the region of the first learning value map 54, that is, in the O2 feedback region. In the plurality of second learning regions B obtained by dividing the lean region into four, the width of the engine speed NE is set to be narrower than that of the first learning value map 54.

  The plurality of second learning areas B include a second learning area B set in the first learning area A2 (hereinafter referred to as a second learning area B2) and a second learning area B set in the first learning area A3. (Hereinafter referred to as the second learning region B3), the second learning region B set in the first learning region A4 (hereinafter referred to as the second learning region B4), and the second learning set in the first learning region A5. Region B (hereinafter, second learning region B5).

  In the second learning area B2, the width of the engine speed NE is narrower than the corresponding first learning area A2, and the throttle opening TH is equal to that of the first learning area A2 when the engine speed NE is low. Although it is the same as a width | variety, it becomes narrow with respect to the width | variety of 1st learning area | region A2 as it becomes a high rotation side. Specifically, in the second learning region B2, the width of the engine speed NE increases as the throttle opening TH increases, and the lower limit of the throttle opening TH increases as the engine speed NE increases. . In the second learning region B3, the width of the engine speed NE is narrower than the corresponding first learning region A3, and the throttle opening TH is the same width as the first learning region A3. In the second learning region B4, the width of the engine speed NE is narrower than the corresponding first learning region A4, and the throttle opening TH is the same width as the first learning region A4. In the second learning region B5, the width of the engine speed NE is narrower than the corresponding first learning region A5, and the throttle opening TH is also narrower than the first learning region A5. Specifically, in the second learning area B5, the upper limit of the throttle opening TH is set lower than that in the first learning area A5. The second learning area B2 is the second learning area B where the throttle opening TH is on the low opening side, and the second learning area B5 is the second learning area B where the throttle opening TH is on the high opening side.

  FIG. 7 is a block diagram illustrating a configuration of the control unit 42. The control unit 42 includes a basic injection amount calculation unit 60, an O2 feedback unit 62, a learning value map 64 having a first learning value map 54 and a second learning value map 56 shown in FIGS. 4 to 6, and a volatile memory. 66, a nonvolatile memory 68, and a fuel injection amount calculation unit 70.

  The basic injection amount calculation unit 60 has an injection amount map 80. The injection amount map 80 is a three-dimensional map in which the basic injection amount TIMB corresponding to the throttle opening degree TH and the engine speed NE is stored. This basic injection amount TIMB is set in advance by simulation or experiment so that the ideal air-fuel ratio can be obtained.

  FIG. 8 is a diagram illustrating an example of the injection amount map 80. FIG. 8 shows the basic injection amount TIMB when the engine speed NE is a predetermined number of times. As shown in FIG. 8, the basic injection amount TIMB increases as the throttle opening TH increases.

  The basic injection amount calculation unit 60 acquires, from the injection amount map 80, a basic injection amount TIMB corresponding to the throttle opening TH detected by the throttle opening sensor 44 and the engine speed NE detected by the engine speed sensor 46. The basic injection amount TIMB is calculated. This basic injection amount TIMB can be expressed by the time during which the fuel injection valve 40 is open.

  The O2 feedback unit 62 calculates a correction coefficient that brings the air-fuel ratio closer to the ideal air-fuel ratio (stoichiometry) when the current throttle opening TH and engine speed NE are within the O2 feedback region. The O2 feedback unit 62 includes a rich lean determination unit 82 and a coefficient calculation unit 84. The rich / lean determination unit 82 determines exhaust gas rich or lean based on the output signal of the oxygen sensor 50.

  The coefficient calculation unit 84 includes a feedback correction amount calculation unit 84a and a learning value calculation unit 84b. When the current throttle opening TH and engine speed NE are within the O2 feedback region, the feedback correction amount calculation unit 84a uses KO2 as a feedback correction amount and its average value based on the determination result of the rich lean determination unit 82. A certain KO2ave is calculated. The calculated KO2 and KO2ave are stored in the volatile memory 66.

  The learning value calculation unit 84b calculates KBU according to KO2 and calculates KBUave (first learning value) when the current throttle opening TH and engine speed NE are within the O2 feedback region. The learning value calculation unit 84b calculates KBULEAN (second learning value) when the current throttle opening degree TH and engine speed NE are within the O2 feedback region and the leaning region. This KBUave is the average value of the KBU for O2 feedback control, and KBULEAN is the average value of the KBU for lean control.

  The calculated KBU, KBUave, and KBULEAN are stored in the volatile memory 66. The nonvolatile memory 68 stores KBULEAN. KO2 and KBUave are correction coefficients used when calculating the fuel injection amount TOUT during the O2 feedback control, and are used to bring the air-fuel ratio closer to the ideal air-fuel ratio (stoichiometric).

  As shown in FIG. 9, the volatile memory 66 includes a KO2 storage unit 90 that stores KO2 and a KO2ave storage unit 92 that stores KO2ave. Further, the volatile memory 66 includes a KBU1 storage unit 94, a KBU2 storage unit 96, a KBU3 storage unit 98, and a KBU4 storage unit 100 that store the KBUs (hereinafter referred to as KBU1 to KBU6) of the first learning region A1 to the first learning region A6. , A KBU5 storage unit 102, and a KBU6 storage unit 104.

  The volatile memory 66 includes a KBUave1 storage unit 106, a KBUave2 storage unit 108, a KBUave3 storage unit 110, and a KBUave4 storage unit 112 that store KBUave (hereinafter referred to as KBUave1 to KBUave6) of the first learning region A1 to the first learning region A6. , A KBUave5 storage unit 114, and a KBUave6 storage unit 116.

  Further, the volatile memory 66 includes a KBULEAN2 storage unit 118, a KBULEAN3 storage unit 120, a KBULEAN4 storage unit 122, and a KBULEAN5 storage unit that store KBULEAN of the second learning region B2 to the second learning region B5 (hereinafter referred to as KBULEAN2 to KBULEAN5). 124. Only the latest values are stored in the storage units (90 to 124).

  The feedback correction amount calculation unit 84a calculates KO2 when the determination result of the rich / lean determination unit 82 is switched (from rich to lean, or from lean to rich). The feedback correction amount calculating unit 84a updates KO2 by overwriting and storing the calculated KO2 in the KO2 storage unit 90 of the volatile memory 66. Note that when the external environment (temperature, pressure, humidity, etc.) and the operating state are constant and the air-fuel ratio is in the vicinity of the ideal air-fuel ratio, the determination result of the rich lean determination unit 82 switches approximately periodically.

  Further, the feedback correction amount calculation unit 84a calculates KO2ave of KO2 calculated up to a predetermined number of times before the calculation timing of KO2 (for example, KO2 for the past three times in total, previous and previous times). The feedback correction amount calculation unit 84a updates the KO2ave every time KO2 is calculated by overwriting the calculated KO2ave in the KO2ave storage unit 92 of the volatile memory 66 and storing it.

  The learning value calculation unit 84b uses the first learning value map 54 to detect the throttle opening TH detected by the throttle opening sensor 44 and the engine speed sensor 46 when the determination result of the rich lean determination unit 82 is switched. The first learning area A specified by the engine speed NE is referred to, and the KBU (any one of KBU1 to KBU6) of the referred first learning area A is calculated and updated.

  The calculation of the KBU is performed in the first learning area A specified by the fluctuation (KO2ave / reference value) with respect to the KO2ave reference value stored in the KO2ave storage unit 92, the current throttle opening TH, and the engine speed NE. Is multiplied by the current KBU to calculate a new KBU. The learned value calculation unit 84b stores the calculated KBU in the volatile memory 66, so that the first learning area A specified by the throttle opening TH and the engine speed NE at that time is calculated each time KO2 is calculated. Update the KBU.

  For example, when the first learning region A specified by the current throttle opening TH and the engine speed NE is the first learning region A2, the learning value calculation unit 84b stores the KO2ave stored in the KO2ave storage unit 92. A new KBU2 is calculated by multiplying the fluctuation (KO2ave / reference value) with respect to the reference value by the KBU2 stored in the KBU2 storage unit 96. The newly calculated KBU2 is overwritten and stored in the KBU2 storage unit 96.

  When the learning value calculation unit 84b calculates the KBU of the first learning region A specified by the current throttle opening TH and the engine speed NE, the learning value calculation unit 84b newly calculates KBUave, which is the average value of the KBUs of the first learning region A. calculate. This KBUave is obtained by calculating the average value of the KBUs of the same first learning area A calculated up to a predetermined number of times before (for example, the average value of the previous five times of the BUBUs up to the previous four times). it can. The learned value calculation unit 84 b stores the calculated KBUave in the volatile memory 66. Thus, every time KBU is calculated, the KBUave of the first learning area A specified by the throttle opening TH and the engine speed NE at that time is updated.

  For example, when the first learning area A corresponding to the current driving area is the first learning area A2, the learning value calculation unit 84b calculates KBUave2, which is an average value of KBU2 for the past predetermined number of times in the first learning area A2. The calculated KBUave2 is overwritten and stored in the KBUave2 storage unit 108.

  When the current throttle opening TH and the engine speed NE are within the second learning region B set within the first learning region A, the learning value calculation unit 84b calculates the average value of the KBUs in the second learning region B. Calculate KBULEAN. That is, KBULEAN, which is an average value of KBUs calculated when the throttle opening TH and the engine speed NE are within the second learning region B, is calculated. This KBULEAN is a KBU calculated when the throttle opening degree TH and the engine speed NE are within the second learning area B, and the KBU within the same second learning area B calculated up to a predetermined number of times (for example, In this case, the average value of the previous three times (KBU for the past three times in the past and the previous time) can be obtained. The learned value calculation unit 84 b stores the calculated KBULEAN in the volatile memory 66. Thus, whenever KBU is calculated when the throttle opening TH and the engine speed NE are within the second learning region B, the KBULEAN of the second learning region B at that time is updated.

  For example, when the current throttle opening TH and the engine speed NE are within the first learning area A2 and the second learning area B2, the learned value calculation unit 84b determines the throttle opening TH and the engine speed. Calculate KBULEAN2, which is an average value of KBU2 for the past predetermined number of times, calculated when NE is in the second learning region B2. The calculated KBULEAN2 storage unit 118 stores the result.

  When KBULEAN is updated, since the throttle opening TH and the engine speed NE are naturally within the O2 feedback region, KBULave is also updated. That is, when the throttle opening TH and the engine speed NE are in the first learning area A and in the second learning area B, the KBU and KBUave of the first learning area A are updated. The KBULEAN of the second learning area B is also updated. When the throttle opening TH and the engine speed NE are in the first learning area A and not in the second learning area B, only KBU and KBUave in the first learning area A are updated.

The fuel injection amount calculation unit 70 calculates the fuel injection amount TOUT injected from the fuel injection valve 40. Specifically, the fuel injection amount calculation unit 70 calculates the fuel according to the equation (1) when the throttle opening TH and the engine speed NE are within the O2 feedback region and the lean operation is not performed. An injection amount TOUT is calculated.
TOUT = TIMB × KO2 × KBUave × KTA (1)

  Here, TIMB in the equation (1) is the basic injection amount TIMB calculated by the basic injection amount calculation unit 60 according to the current throttle opening TH and the engine speed NE, and KO2 is stored in the KO2 storage unit 90. KO2. The KBUave in the equation (1) is the KBUave of the first learning area A specified by the current throttle opening TH and the engine speed NE, and is stored in the volatile memory 66. For example, when the current throttle opening TH and the engine speed NE are within the first learning area A3, they are KBUave3 stored in the KBUave3 storage unit 110. KTA in equation (1) is an intake air temperature correction coefficient KTA corresponding to the intake air temperature TA detected by the intake air temperature sensor 52. The intake air temperature correction coefficient KTA is acquired from a table or map (not shown). Note that the fuel injection amount TOUT may be calculated using KO2ave as a feedback correction amount instead of KO2 in (1).

When the throttle opening TH and the engine speed NE are within the leaning range during the leaning mode in which the leaning operation described later can be performed, the fuel injection amount calculation unit 70 performs the leaning control according to the equation (2). A fuel injection amount TOUT is calculated. Thereby, the lean operation can be performed.
TOUT = TIMB × KO2s × KBULEAN × KTA × KLEAN (2)

  TIMB and KTA in equation (2) are the same as in equation (1). The KO2s in the equation (2) is a standard value of KO2 used instead when the KO2 cannot be calculated as in the lean operation, and is a predetermined coefficient. This KO2s is stored in a memory (not shown). KBULEAN in equation (2) is KBULEAN in the second learning region B specified by the current throttle opening TH and engine speed NE, and is stored in the volatile memory 66. For example, when the current throttle opening TH and the engine speed NE are within the second learning region B3, it is KBULEAN3 stored in the KBULEAN3 storage unit 120. In the equation (2), KLEAN is a lean control coefficient that varies according to the throttle opening TH and the engine speed NE, and is a coefficient (e.g., 0.8) for thinning the fuel.

  In the case of leaning control in which O2 feedback control cannot be performed, KO2, its KO2ave, and KBU, KBUave, and KBULEAN are not updated sequentially, so during leaning control, KBULEAN before entering leaning control is used as it is. Will be.

  The fuel injection amount TOUT can be represented by the time during which the fuel injection valve 40 is open. When the fuel injection valve 40 is opened for the time represented by the calculated fuel injection amount TOUT, the fuel is injected from the fuel injection valve 40. Be injected.

  O2 feedback control means that KO2, KO2ave, KBU, and KBUave are updated when the throttle opening TH and the engine speed NE are within the O2 feedback region and the lean operation is not performed. This means a series of controls for performing the process of calculating the fuel injection amount TOUT using the updated KO2 and KO2ave. KBULEAN is also updated during this O2 feedback control except during the lean mode described later.

  Thus, in the present embodiment, separately from the first learning value map 54, it is set in the first learning value map 54, and the width of the engine speed NE is set lower than that in the first learning value map 54. A second learning value map 56 is provided. The width of the engine speed NE of each second learning area B of the second learning value map 56 of the second learning value map 56 is set narrower than the first learning area A of the corresponding first learning value map 54. Yes.

  Here, even for KBUave within the same first learning region A, there is a possibility that a large difference may occur in the value of KBUave calculated when the engine speed NE is high and low. . For example, in the first learning area A3, KBUave by KBU learned at a point indicated by P31 in FIG. 5 (the engine speed NE is on the low speed side) and a point indicated by P32 (the engine speed NE is high). In many cases, there is a difference from KBUave by KBU learned on the rotation side.

  However, since the width of the engine speed NE in the second learning area B is narrower than that in the first learning area A, the engine speed NE in the same second learning area B is calculated on the high speed side and the low speed side. The difference in KBULEAN can be made smaller than in the case of KBUave, and the accuracy of KBULEAN can be increased.

  During the O2 feedback control, even if the operation region (for example, P31) in which KBU has been learned before is different from the operation region (for example, P32) at the current fuel injection amount calculation timing, the calculation of KO2 Since KBUave and the like are updated sequentially each time, the influence on the fuel injection amount TOUT calculated by the above equation (1) is small. On the other hand, in the case of lean control, KBUave, KBULEAN, etc. are not updated. For this reason, when the KBUave of the first learning value map 54 stored in the volatile memory 66 is calculated on the low rotation side, the volatile memory 66 when the current engine speed NE is on the high rotation side. There is a possibility that the deviation from the KBUave stored in the table becomes large.

  Therefore, during lean control, KBULEAN of the second learning value map 56 is used instead of KBUave of the first learning value map 54. Thereby, even if the current engine speed NE is on the high speed side when KBULEAN of the second learning value map 56 stored in the volatile memory 66 is calculated on the low speed side, Deviation from KBUave stored in the volatile memory 66 can be reduced. As a result, an appropriate fuel injection amount TOUT can be calculated.

  Regarding the first learning region A5, there may be a relatively large difference in the value of KBUave calculated between the low opening side and the high opening side of the accelerator opening. For example, KBUave by KBU learned at the point indicated by P51 (throttle opening TH is on the low opening side) in FIG. 5 and learning at the point indicated by P52 (throttle opening TH is on the high opening side). In many cases, there is a difference in KBUave by KBU. Therefore, in the second learning area B5, the upper limit of the throttle opening TH is set lower than that in the first learning area A5. As a result, the difference in KBULEAN5 calculated when the throttle opening TH in the same second learning region B5 is high and low can be reduced as compared to the case of KBUave5, and the accuracy of KBULEAN5 Can be high.

  FIG. 10 is a flowchart showing the setting operation of the leaning mode in which the leaning operation is possible. First, the control unit 42 determines whether or not the water temperature TW detected by the water temperature sensor 48 is equal to or higher than a predetermined temperature (step S1).

  If it is determined in step S1 that the water temperature TW is equal to or higher than the predetermined temperature, the control unit 42 determines whether or not the vehicle on which the fuel injection control device 12 is mounted is in a neutral state (step S2).

  If it is determined in step S2 that it is in the neutral state, it is determined whether KBUave1 and KBULEAN3 and 4 have been learned, that is, whether or not they are stored in the volatile memory 66 (step S3).

  If it is determined in step S3 that KBUave1 and KBULEAN3 and 4 have been learned, the control unit 42 sets the lean mode (step S4). If it is determined in step S1 that the water temperature TW is lower than the predetermined temperature, if it is determined in step S2 that it is in a neutral state, it is determined in step S3 that at least one of KBUave1 and KBULEAN3 and 4 is not learned. If not, do not set to lean mode.

  FIG. 11 is a flowchart showing the operation of the fuel injection control device 12 when the throttle opening TH and the engine speed NE are within the O2 feedback region. First, the control unit 42 determines whether or not it is currently in the lean mode (step S11).

  If it is determined in step S11 that the leaning mode is in effect, the control unit 42 determines that the current throttle opening TH and engine speed NE detected by the throttle opening sensor 44 and the engine speed sensor 46 are within the leaning range. It is determined whether or not (step S12).

  If it is determined in step S12 that the current throttle opening TH and the engine speed NE are within the lean range, the control unit 42 determines the second learning area B specified by the throttle opening TH and the engine speed NE. It is determined whether or not KBULEAN is stored in the volatile memory 66 (whether it has been learned) (step S13). For example, when the second learning area B specified by the throttle opening TH and the engine speed NE is the second learning area B5, it is determined whether or not KBULEAN5 has been learned.

  If it is determined in step S11 that the current mode is not in the leaning mode, if it is determined in step S12 that the current throttle opening TH and the engine speed NE are not within the leaning range, the current throttle opening is determined in step S13. If it is determined that KBULEAN of the second learning region B specified by the degree TH and the engine speed NE has not been learned, the process proceeds to step S14, and the control unit 42 performs O2 feedback control.

  Specifically, the fuel injection amount calculation unit 70 acquires from the volatile memory 66 the KBUave and KO2 of the first learning region A specified by the current throttle opening TH and the engine speed NE. Then, the basic injection amount calculation unit 60 adds the basic injection amount TIMB calculated according to the current throttle opening TH and the engine speed NE to the intake air temperature correction according to the acquired KO2 and KBUave and the current intake air temperature TA of KO2. The fuel injection amount TOUT is calculated by multiplying the coefficient KTA. That is, the fuel injection amount TOUT is calculated using the above-described equation (1). The fuel injection valve 40 injects fuel with the calculated fuel injection amount TOUT.

  On the other hand, when it is determined in step S13 that KBULEAN of the second learning region B specified by the current throttle opening TH and engine speed NE has been learned, the control unit 42 performs leaning control to make leaning Operation is performed (step S15).

  Specifically, the fuel injection amount calculation unit 70 acquires KBULEAN of the second learning region B according to the current throttle opening TH and engine speed NE from the volatile memory 66. Then, the basic injection amount calculation unit 60 adds the basic injection amount TIMB calculated according to the current throttle opening TH and the engine speed NE to KO2s, the acquired KBULEAN, and the intake air temperature correction according to the current intake air temperature TA. The fuel injection amount TOUT is calculated by multiplying the coefficient KTA and the lean control coefficient KLEAN. That is, the fuel injection amount TOUT is calculated using the above-described equation (2). The fuel injection valve 40 injects fuel with the calculated fuel injection amount TOUT. The fuel injection amount TOUT calculated by the equation (2) is less fuel than the fuel injection amount TOUT calculated by the equation (1), and the lean amount in which the fuel injected from the fuel injection valve 40 is reduced. Operation can be performed.

  The coefficient calculator 84 sequentially calculates at least KO2, KO2ave, KBU, and KBUave during the O2 feedback control, and the latest value is stored in the volatile memory 66. This update of KBUave and the like will be described with reference to the flowchart of FIG.

  First, the coefficient calculation unit 84 determines whether or not the calculation timing of KO2 has arrived (step S21). The coefficient calculation unit 84 determines that the calculation timing of KO2 has arrived when the determination result of the rich lean determination unit 82 is switched (from rich to lean, or from lean to rich).

  If it is determined in step S21 that the calculation timing of KO2 has not arrived, the process stays in step S21 until it arrives. If it is determined that the calculation timing has arrived, the process proceeds to step S22. In step S22, the feedback correction amount calculation unit 84a of the coefficient calculation unit 84 calculates and updates KO2 and KO2ave. That is, the calculated KO2 is overwritten in the KO2 storage unit 90 of the volatile memory 66, and the calculated KO2ave is overwritten in the KO2ave storage unit 92 of the volatile memory 66.

  Next, the learning value calculation unit 84b specifies the first learning region A based on the current throttle opening TH and the engine speed NE detected by the throttle opening sensor 44 and the engine speed sensor 46 (step S23). In other words, the first learning region A specified by the current throttle opening TH and the engine speed NE is determined.

  Next, the learning value calculation unit 84b calculates and updates the KBU and KBUave of the specified first learning region A (step S24). That is, the calculated KBU and KBUave of the first learning area A are overwritten in the volatile memory 66. For example, when the specified first learning area A is the first learning area A5, KBU5 and KBUave5 are calculated, and the calculated KBU5 and KBUave5 are overwritten on the KBU5 storage unit 102 and the KBUave5 storage unit 114 of the volatile memory 66. Is done.

  Next, the learning value calculation unit 84b determines whether or not the current throttle opening TH and the engine speed NE are within the second learning region B set within the first learning region A identified at step S23. Judgment is made (step S25). For example, when the specified first learning area A is the first learning area A5, it is determined whether or not the current throttle opening TH and the engine speed NE are within the second learning area B5.

  If it is determined in step S25 that the throttle opening TH and the engine speed NE are within the second learning region B set in the first learning region A specified in step S23, the learned value calculation unit 84b The KBULEAN of the second learning area B is calculated and updated (step S26). That is, the calculated KBULEAN of the second learning area B is overwritten in the volatile memory 66. For example, when it is determined that the current throttle opening TH and the engine speed NE are within the second learning region B5, KBULEAN5 is calculated, and the calculated KBULEAN5 is overwritten in the KBULEAN5 storage unit 124 of the volatile memory 66. The

  On the other hand, if it is determined in step S25 that the throttle opening TH and the engine speed NE are not within the second learning area B set in the first learning area A determined in step S23, KBULEAN is updated. The process ends without being executed.

  As described above, when the throttle opening TH and the engine speed NE are within the O2 feedback region in the case where the lean mode is not being performed, the O2 feedback control is executed. As a result, KO2 and KO2ave are updated, and the KBU and KBUave of each first learning region A are updated. Further, when the throttle opening TH and the engine speed NE enter the lean range during the O2 feedback control, the KBULEAN of the second learning area B specified by the throttle opening TH and the engine speed NE further becomes Updated.

  Even in the lean mode, if the throttle opening TH and the engine speed NE are not within the lean range, O2 feedback control is executed. As a result, KO2 and KO2ave are updated, and the KBU and KBUave of each first learning region A are updated.

  Even in the lean mode, even when the throttle opening TH and the engine speed NE are within the lean range, KBULEAN in the second learning area B specified by the throttle opening TH and the engine speed NE. Is not learned, O2 feedback control is executed. Thereby, in addition to KO2 and KO2ave, KBU and KBUave in the first learning area A are updated, and KBULEAN in the second learning area B that has not been learned is also learned.

  From the above, when entering the lean mode, in principle, KBULEAN in the second learning region B is not updated, and only KBULEAN in the second learning region B that has not been learned as an exception is learned.

  The lean mode is canceled when the change rate ΔTH of the throttle opening TH exceeds the threshold, when the water temperature TW falls below a predetermined value, or when fail control is entered due to a malfunction such as a sensor failure.

  As described above, in the present embodiment, during O2 feedback control, the KBUave of each first learning region A in the first learning value map 54 is updated to calculate the fuel injection amount TOUT. Equivalent KBUave learning speed and O2 feedback control accuracy can be ensured. In addition, when the second learning value map 56 used for the lean operation is set in the region of the first learning value map 54 and learning is performed in the second learning value map 56 during the O2 feedback control, the first learning value map 56 is used. By learning both KBUave of the first learning area A in the value map 54 and KBULEAN of the second learning area B in the second learning value map 56, the same throttle opening TH and engine speed NE are obtained. However, different learning values can be learned. Therefore, the accuracy of KBULEAN used for the predetermined operation is improved, and when performing the predetermined operation within the region of the second learning value map 56, high-accuracy fuel injection is achieved by using KBULEAN of the second learning value map 56. It can be performed.

  When entering the lean mode, updating (learning) of KBULEAN of each second learning region B of the second learning value map 56 before entering the lean mode is stopped, so it is assumed during the lean operation. Deviation from the air-fuel ratio can be suppressed.

  When the leaning operation is not executed during the leaning mode, the update (learning) of the KBUave is performed, so that the accuracy of the KBUave can be ensured even during the leaning mode.

  The second learning region B2 in which the throttle opening TH of the second learning value map 56 is on the low opening side is set such that the width of the engine speed NE increases as the throttle opening TH increases. The range of possible operation can be expanded as much as possible to reduce fuel consumption.

  The second learning area B5 where the throttle opening TH of the second learning value map 56 is on the higher opening side is more throttle opening than the first learning area A5 of the first learning value map 54 corresponding to the second learning area B5. Since the upper limit of TH is set low, the region in which the lean operation can be performed can be appropriately limited, and the operation fluctuation of the engine 10 during the lean operation can be suppressed.

  The nonvolatile memory 68 has the first calculated values of KBUave1 to 5, the values of KBUave1 to 6 stored in the volatile memory 66 immediately before the leaning mode is set, or the leaning mode. The values of KBUave1 to KBUave6 calculated first after the cancellation are stored. The KBUaves 1 to 6 stored in the nonvolatile memory 68 are used for calculating the fuel injection amount TOUT during the O2 feedback control immediately after the engine is started.

DESCRIPTION OF SYMBOLS 10 ... Engine 12 ... Fuel injection control apparatus 22 ... Intake pipe 24 ... Exhaust pipe 30 ... Intake valve 32 ... Exhaust valve 34 ... Throttle valve 40 ... Fuel injection valve 42 ... Control part 44 ... Throttle opening sensor 46 ... Engine speed sensor DESCRIPTION OF SYMBOLS 50 ... Oxygen sensor 54 ... 1st learning value map 56 ... 2nd learning value map 60 ... Basic injection amount calculation part 62 ... Feedback part 64 ... Learning value map 66 ... Volatile memory 70 ... Fuel injection amount calculation part 80 ... Injection amount Map 82 ... Rich lean determination unit 84 ... Coefficient calculation unit 84a ... Feedback correction amount calculation unit 84b ... Learning value calculation unit

Claims (5)

An oxygen sensor (50) for detecting the oxygen concentration in the exhaust gas of the engine (10);
When the throttle opening (TH) of the throttle valve (34) provided in the intake pipe (22) of the engine (10) and the engine speed (NE) are within the O2 feedback region, the oxygen sensor (50 ) Calculates a feedback correction amount (KO2) that is a correction coefficient based on the detected oxygen concentration, and calculates a first learning value (KBUave) that is a correction coefficient learned according to the feedback correction amount (KO2). A plurality of first learning value maps (54) represented by a throttle opening (TH) and the engine speed (NE) and having the O2 feedback region divided according to the throttle opening (TH). A coefficient calculation unit (84) that calculates and updates for each learning area (A);
When the throttle opening (TH) and the engine speed (NE) are within the O2 feedback region, the basic injection amount (TIMB) is set to the feedback correction amount (KO2) and the first learning value (KBUave). A fuel injection amount calculation unit (70) that calculates the fuel injection amount (TOUT) by correcting using
In a fuel injection control device (12) of an engine (10) comprising:
A plurality of second learning regions that are set within the region of the first learning value map (54) and at least the width of the engine speed (NE) is set narrower than that of the first learning value map (54) A second learning value map (56) having (B),
When the throttle opening (TH) and the engine speed (NE) are within the second learning value map (56), the coefficient calculation unit (84) further includes the throttle opening (TH). And the second learning value (KBULEAN) of the second learning region (B) of the second learning value map (56) corresponding to the engine speed (NE) and the engine speed (NE) according to the feedback correction amount (KO2). Updated,
The fuel injection amount calculation unit (70) performs a predetermined operation when the throttle opening (TH) and the engine speed (NE) are within the range of the second learning value map (56). An engine (10) that calculates a fuel injection amount (TOUT) by correcting the basic injection amount (TIMB) using the second learning value (KBULEAN) instead of the first learning value (KBUave). ) Fuel injection control device (12). In the fuel injection control device (12) of the engine (10) according to claim 1,
The predetermined operation is a lean operation,
The coefficient calculation unit (84) stops updating the second learning value (KBULEAN) when entering the lean mode in which the lean operation is possible. The fuel injection control device (10) for the engine (10), 12). In the fuel injection control device (12) of the engine (10) according to claim 2,
When the throttle opening (TH) and the engine speed (NE) are within the range of the second learning value map (56) during the lean mode, the lean operation is executed.
The coefficient calculation unit (84) updates the first learning value (KBUave) when the leaning operation is not executed, and the first learning value (KBUave) when the leaning operation is executed. The fuel injection control device (12) of the engine (10) is characterized by stopping the update of the engine. In the fuel injection control device (12) of the engine (10) according to any one of claims 1 to 3,
In the second learning region (B2) where the throttle opening (TH) of the second learning value map (56) is low, the engine speed (NE) increases as the throttle opening (TH) increases. The fuel injection control device (12) of the engine (10) is characterized in that it is set so as to widen. In the fuel injection control device (12) of the engine (10) according to any one of claims 1 to 4,
The second learning region (B5) in which the throttle opening (TH) of the second learning value map (56) is on the higher opening side is the first learning value map (B5) corresponding to the second learning region (B5). 54) The fuel injection control device (12) of the engine (10), wherein the upper limit of the throttle opening (TH) is set lower than the first learning region (A5) of 54).

Images