JP2015140791A - Control device of internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To implement learning of a composition of a gas fuel without affected by change of a fuel system.SOLUTION: A fuel supply system includes: a gas tank 42 for storing a gas fuel in a high pressure state; and a first injection valve 21 as gas injection means for injecting the gas fuel supplied from the gas tank 42 through a fuel passage. A control portion 80 changes an ignition timing of an engine 10 until an actual torque reaches a prescribed target torque at the side to eliminate torque change, when occurrence of torque change accompanying replenishment of the gas fuel into the gas tank 42 is detected on the actual torque generated in the engine 10. Learning of a composition of the gas fuel supplied to the first injection valve 21 is implemented on the basis of a change amount of the ignition timing at that time.

Description

  The present invention relates to a control device for an internal combustion engine, and more particularly to a control device for an internal combustion engine operable by combustion of gaseous fuel.

  Conventionally, vehicles equipped with an internal combustion engine that is driven by burning gaseous fuel such as compressed natural gas (CNG) have been put into practical use. In gaseous fuels such as CNG, the fuel composition varies depending on the production area and production process. Further, the amount of fuel required for each combustion differs depending on the fuel composition. For example, when the amount of inert gas (impurities) in the gaseous fuel increases, the amount of fuel required for each combustion increases. On the other hand, when an appropriate amount of fuel according to the fuel composition is not injected, it is conceivable that knocking or misfire may occur due to poor combustion of the engine. In view of this, various techniques have been proposed in the past for improving combustion failures caused by variations in fuel composition in internal combustion engines that use gaseous fuel (see, for example, Patent Document 1).

  In the control device described in Patent Document 1, whether or not the fuel composition of the gaseous fuel has changed is determined based on whether or not the state where the engine rotational speed is equal to or lower than a predetermined value during engine startup continues for a predetermined period. Alternatively, the determination is made based on whether or not the state in which the throttle opening is equal to or greater than a predetermined value at the time of engine start continues for a predetermined period. Further, when it is determined that the fuel composition of the gaseous fuel has changed, the learning value of the air-fuel ratio feedback control that has been used so far is deleted and learned again. Further, the control device described in Patent Document 1 estimates the octane number of the fuel by obtaining the deviation of the theoretical air-fuel ratio from the air-fuel ratio correction value. Further, the ignition timing correction value is obtained from the estimated octane number, and the basic ignition timing is corrected with the ignition timing correction value, thereby calculating the final ignition timing according to the fuel composition.

JP 2000-170581 A

  For example, when an air-fuel ratio shift occurs due to a fuel system such as aging deterioration or purging of the fuel injection valve, the influence appears in the air-fuel ratio correction value as in the case of the variation in fuel composition. Therefore, when trying to learn the fuel composition based on the air-fuel ratio correction value, the learning accuracy of the fuel composition learning is reduced by recognizing that the air-fuel ratio deviation due to factors other than the variation in fuel composition is due to the difference in fuel composition. Can be considered.

  The present invention has been made to solve the above-described problems, and has as its main object to provide a control device for an internal combustion engine that can perform composition learning of gaseous fuel without being affected by changes in the fuel system. To do.

  The present invention employs the following means in order to solve the above problems.

  The present invention comprises an internal combustion engine (10) comprising a fuel tank (42) for storing gaseous fuel in a high pressure state, and gas injection means (21) for injecting gaseous fuel supplied from the fuel tank through a fuel passage (41). The present invention relates to a control device for an internal combustion engine applied to. According to a first aspect of the present invention, there is provided a change detecting means for detecting a change in torque associated with the replenishment of the gaseous fuel into the fuel tank with respect to the actual torque generated in the internal combustion engine, and the change detecting means. Ignition control means for changing the ignition timing of the internal combustion engine until the actual torque reaches a predetermined target torque on the side of canceling the torque change, when the occurrence of the torque change is detected by Learning execution means for performing composition learning of the gaseous fuel supplied to the gas injection means based on the amount of change in the ignition timing by the control means.

  In the above configuration, paying attention to the fact that the relationship between the actual torque generated by combustion in the internal combustion engine and the ignition timing differs depending on the fuel composition (for example, the mixing ratio of the inert gas), the gaseous fuel into the fuel tank When it is detected that a torque change due to replenishment has occurred, the ignition timing is changed until the actual torque matches the target torque determined on the side that cancels the torque change, and the fuel is changed based on the amount of change in the ignition timing. Perform composition learning. That is, the influence of the fuel composition deviation is absorbed by the change amount of the ignition timing, and the fuel composition is learned based on the magnitude of the change amount of the ignition timing at that time. According to such a configuration, since the fuel composition learning is performed regardless of the parameters of the fuel system, the fuel composition learning can be performed without being affected by changes in the fuel system.

1 is an overall schematic configuration diagram of an engine control system. FIG. The torque characteristic figure which shows the relationship between ignition timing and engine torque. The flowchart which shows the process sequence of the fuel composition learning process of 1st Embodiment. The figure which shows an example of a combustion speed estimation map. The figure which shows an example of an inert gas calculation map. The flowchart which shows the process sequence of the abnormality diagnosis process of an intake pressure sensor. The flowchart which shows the process sequence of a fuel injection amount / ignition timing correction process. The flowchart which shows the process sequence of the fuel composition learning process of 2nd Embodiment.

(First embodiment)
The first embodiment will be described below with reference to the drawings. The present embodiment is an engine control system applied to a so-called bi-fuel type in-vehicle engine (internal combustion engine) that uses compressed natural gas (CNG) as a gaseous fuel and gasoline as a liquid fuel as combustion fuel. It is materialized. An overall schematic diagram of this system is shown in FIG.

  An engine 10 shown in FIG. 1 is a multi-cylinder (for example, in-line three-cylinder) spark ignition engine. An intake pipe 11 is connected to the intake port of the engine 10 via an intake manifold 12, and an exhaust pipe 14 is connected to the exhaust port via an exhaust manifold 13. The intake pipe 11 is provided with a throttle valve 15 as air amount adjusting means. The throttle valve 15 is configured as an electronically controlled throttle valve whose opening is adjusted by a throttle actuator 15a such as a DC motor. The opening degree of the throttle valve 15 (throttle opening degree) is detected by a throttle opening degree sensor 15b built in the throttle actuator 15a.

  The exhaust pipe 14 is provided with an exhaust sensor for detecting exhaust components and a catalyst 19 for purifying the exhaust. As the exhaust sensors, oxygen sensors 18a and 18b that output detection signals corresponding to the oxygen concentration in the exhaust are provided on the upstream side and the downstream side of the catalyst 19, respectively.

  An intake valve 25 and an exhaust valve 26 as engine valves for adjusting the amount of air introduced into the cylinder are respectively provided at the intake port and the exhaust port of the engine 10. When the intake valve 25 is opened, a mixture of air and fuel is introduced into the cylinder, and when the exhaust valve 26 is opened, the exhaust gas after combustion is discharged into the exhaust passage.

  A spark plug 20 is provided in each cylinder of the engine 10. A high voltage is applied to the ignition plug 20 at a desired ignition timing through an ignition device 20a including an ignition coil. By applying this high voltage, a spark discharge is generated between the opposing electrodes of each spark plug 20, and the fuel introduced into the cylinder is ignited and used for combustion.

  This system is provided with a first injection valve 21 for injecting gaseous fuel and a second injection valve 22 for injecting liquid fuel as fuel injection means for injecting and supplying fuel to each cylinder of the engine 10. Yes. These injection valves 21 and 22 respectively inject fuel into the intake manifold 12.

  Each of the injection valves 21 and 22 is an open / close type control valve in which the valve body is lifted from the closed position to the open position by electrically driving the electromagnetic drive unit. Each valve is driven to open by a valve opening drive signal. Each of these injection valves 21 and 22 opens by energization and closes by energization interruption, thereby injecting an amount of fuel corresponding to the energization time. In the present embodiment, the injection pipe 23 is connected to the distal end portion of the first injection valve 21, and the gaseous fuel injected from the first injection valve 21 is injected into the intake manifold 12 through the injection pipe 23. It has become. The first injection valve 21 may be directly attached to the intake pipe 11. The second injection valve 22 is a port injection type in the present embodiment, but may be a direct injection type that directly injects fuel into the cylinder of the engine 10.

  Next, the gas fuel supply unit 40 that supplies gas fuel to the first injection valve 21 and the liquid fuel supply unit 70 that supplies liquid fuel to the second injection valve 22 will be described.

  The gaseous fuel supply unit 40 is provided with a gas tank 42 that stores gaseous fuel in a high pressure state, and a gas pipe 41 that connects the gas tank 42 and the first injection valve 21. In the middle of the gas pipe 41, a regulator 43 is provided as pressure adjusting means having a function of adjusting the pressure of the gaseous fuel supplied to the first injection valve 21 under reduced pressure. The regulator 43 adjusts the pressure of the gaseous fuel in the high-pressure state (for example, a maximum of 20 MPa) stored in the gas tank 42 so as to become a predetermined set pressure (for example, a constant pressure within a range of 0.2 to 1.0 MPa). It is. The gaseous fuel after the decompression adjustment is supplied to the first injection valve 21 through the gas pipe 41. The regulator 43 may be a variable fuel pressure type that can variably adjust the pressure of the fuel supplied to the first injection valve 21.

  The gas pipe 41 further includes a tank main stop valve 44 disposed near the fuel outlet of the gas tank 42, and a shut-off valve disposed downstream of the tank main stop valve 44 and near the fuel inlet of the regulator 43. 45 is provided. These valves 44 and 45 allow and block the flow of gaseous fuel in the gas pipe 41. Both the tank main stop valve 44 and the shut-off valve 45 are electromagnetic on-off valves, and are normally closed types in which the flow of gaseous fuel is blocked when not energized and the flow of gaseous fuel is allowed when energized. In the gas pipe 41, pressure sensors 46 and 47 for detecting fuel pressure are provided upstream and downstream of the regulator 43, and a temperature sensor 48 for detecting fuel temperature is provided downstream of the regulator 43. Is provided.

  The liquid fuel supply unit 70 is provided with a fuel tank 72 that stores liquid fuel, and the fuel tank 72 is connected to the second injection valve 22 via a fuel pipe 71. The fuel pipe 71 is provided with a fuel pump 73 that feeds the liquid fuel in the fuel tank 72 to the second injection valve 22. The liquid fuel pumped up by the fuel pump 73 is supplied to the second injection valve 22 through the fuel pipe 71.

  The control unit 80 includes a CPU, a ROM, a RAM, a backup RAM, and the like, and executes various control programs stored in the ROM, thereby performing various controls of the engine 10 according to each engine operating state. . Specifically, the control unit 80 is electrically connected to the various sensors described above and other sensors (crank angle sensor 81, intake pressure sensor 82, cooling water temperature sensor 83, etc.) provided in the system. The outputs (detection signals) from these sensors are input. The control unit 80 is electrically connected to driving units such as the ignition device 20a and the injection valves 21 and 22, and controls the driving of each driving unit by outputting a driving signal to each driving unit. To do.

  The control unit 80 selectively switches the fuel to be used for the operation of the engine 10 according to the engine operating state, the fuel remaining amount in the tank, an input signal from a fuel selection switch (not shown), and the like. Specifically, when the use of gaseous fuel is selected by the fuel selection switch, or when the remaining amount of liquid fuel in the tank 72 falls below a predetermined value, the operating mode of the engine 10 is set as the gaseous fuel supply unit. A gas fuel mode in which gaseous fuel is supplied to the engine 10 is selected at 40. On the other hand, when the use of the liquid fuel is selected by the fuel selection switch or when the remaining amount of the gaseous fuel in the gas tank 42 falls below a predetermined value, the liquid fuel supply unit 70 sets the liquid as the operation mode of the engine 10. A liquid fuel mode for supplying fuel to the engine 10 is selected.

  Regarding the fuel injection control, the control unit 80 calculates a basic injection amount as a fuel amount sufficient to obtain a target air-fuel ratio (for example, theoretical air-fuel ratio) based on the engine operating state (engine speed and engine load), The final fuel injection amount is calculated by performing various corrections on the injection amount. Examples of various corrections include start-up increase correction, acceleration increase correction, and air-fuel ratio feedback correction. The valve opening period of the fuel injection valve is adjusted so that the calculated fuel injection amount is injected. As a result, a required amount of fuel corresponding to each engine operating state is supplied to the engine 10.

  In this system, torque control is performed so that the actual torque generated by the combustion in the engine 10 becomes the required value. Specifically, the control unit 80 calculates the required torque Trq based on the accelerator opening, the engine rotational speed, and the like, and calculates the actual torque Tra based on the engine operating state (intake air amount, engine rotational speed, etc.). To do. Then, feedback calculation is performed based on the deviation between the actual torque Tra and the required torque Trq, and the intake air amount, ignition timing, fuel injection amount, and the like are calculated based on the result. Instead of calculating (estimating) the actual torque based on the engine operating state, a torque sensor may be provided on the crankshaft of the engine 10 and the actual torque Tra may be calculated based on the detected value of the torque sensor. .

  Here, the composition of CNG varies depending on the production area and production process. There are also areas where mined natural gas is used as fuel without purification. In particular, when the gas tank 42 is filled with a gaseous fuel having a high mixing ratio of an inert gas such as nitrogen or carbon dioxide, the amount of fuel to be injected and the ignition timing for the same amount of air greatly change, and the fuel injection amount and ignition There is a steady shift in time.

  Therefore, in the present embodiment, when the gaseous fuel is replenished in the gas tank 42, the fuel composition of the gaseous fuel supplied to the first injection valve 21 is learned. In particular, in this embodiment, paying attention to the fact that the relationship between the actual torque generated in the engine 10 and the ignition timing differs depending on the fuel composition (the mixing ratio of the inert gas in the fuel), the gaseous fuel into the gas tank 42 When it is detected that a torque change due to replenishment has occurred, the ignition timing is changed until the actual torque reaches a predetermined target torque (maximum torque in the present embodiment) on the side where the torque change is eliminated, The fuel composition learning is executed based on the change amount of the ignition timing at that time.

  The fuel composition learning of this embodiment will be described in detail with reference to the torque characteristic diagram of FIG. In FIG. 2, a curve L1 indicates a gaseous fuel whose fuel composition is a predetermined reference composition, and a curve L2 indicates a gaseous fuel in which the mixing ratio of the inert gas is higher than that of the reference composition.

  Regarding the optimal ignition timing (MBT) that is the ignition timing at which the torque generated in the engine 10 is maximum, the optimal ignition timing (hereinafter also referred to as “reference optimal timing”) when the composition of the gaseous fuel is the reference composition is θm1. Suppose that there is a point (point A in FIG. 2). When gaseous fuel is replenished in the gas tank 42 between the previous engine stop and the current engine start, and the composition of the replenished gaseous fuel is a fuel with a higher mixing ratio of inert gas than the reference composition think about. In this case, when engine combustion is performed with the ignition timing set to the reference optimum timing θm1, the actual torque of the engine 10 is the maximum torque when the composition of the gaseous fuel is the reference composition, as shown by the curve L2 in FIG. The reference torque Tr1 is reduced by ΔTr (point B in FIG. 2). Further, in order to set the actual torque Tra to the maximum torque Tr3 by engine combustion using gaseous fuel after fuel replenishment, it is necessary to advance the ignition timing by Δθig from the reference optimum timing θm1 (point C in FIG. 2). ).

  Focusing on these points, in this embodiment, when the gas fuel is replenished into the gas tank 42, the ignition timing is set to the optimum ignition timing (reference optimum timing θm1) when the composition of the gaseous fuel is the reference composition. It is detected that a torque change caused by the replenishment of the gaseous fuel into the gas tank 42 has occurred based on the torque change when. Then, the ignition timing is changed until the maximum torque (Tr3 in FIG. 2) generated when the gaseous fuel (fuel in the gas tank 42) supplied to the first injection valve 21 is combusted, and the amount of change in the ignition timing at that time. Based on (ignition advance amount Δθig), composition learning of the gaseous fuel supplied to the first injection valve 21 is executed.

  Next, the fuel composition learning process of this embodiment will be described using the flowchart of FIG. This process is performed by the control unit 80 every time the vehicle is driven, for example.

  In FIG. 3, in step S <b> 101, it is determined whether or not the gaseous fuel mode is selected as the operation mode of the engine 10. When the gaseous fuel mode is selected, the process proceeds to step S102, and it is determined whether or not the gaseous fuel has been replenished into the gas tank 42. Whether or not the gaseous fuel has been replenished is determined, for example, by determining whether or not the fuel pressure (tank original pressure) detected by the pressure sensor 46 has become higher than a predetermined value since the previous stop of the engine operation, When the tank original pressure is higher than a predetermined value, it is determined that the fuel is replenished.

  If it is determined that there is replenishment of the gaseous fuel, the process proceeds to step S103, and it is determined whether or not a learning execution condition for fuel composition learning is satisfied. As learning execution conditions, (1) the engine load fluctuation is within a predetermined load fluctuation range and the engine rotation speed is within a predetermined rotation fluctuation range (a predetermined learning operation state); 2) In the present embodiment, the condition that the intake pressure sensor 82 is normal is satisfied. In this embodiment, the learning execution condition is satisfied when all of the conditions including (1) and (2) are satisfied. Is determined. The condition (2) is determined based on the diagnosis result of the abnormality diagnosis process shown in FIG. The case where the condition (1) is satisfied includes an idle operation state and an engine steady state during vehicle travel.

  When it is determined that the learning execution condition is satisfied, the process proceeds to step S104, the torque feedback is stopped, and the engine combustion control for fuel composition learning is switched. In the engine combustion control for fuel composition learning, the control amount immediately before stopping the torque feedback is maintained for the intake air amount and the fuel injection amount, and the ignition timing is set to the reference optimum timing θm1 (point B in FIG. 2). In the subsequent step S105, the actual torque Tra is calculated based on the engine operating state, and a deviation amount of the calculated actual torque Tra from the reference torque Tr1 (torque deviation amount ΔTr = Tr1−Tra) is calculated.

  In step S106, it is determined whether or not the torque deviation amount ΔTr is out of a predetermined allowable range. If the torque deviation amount ΔTr is within the predetermined allowable range, this routine is ended as it is. On the other hand, if the torque deviation amount ΔTr is out of the predetermined allowable range, the process proceeds to step S107. Note that the processing in steps S102 to S106 corresponds to a change detection unit.

  In step S107, the ignition timing is changed by a predetermined amount from the reference optimum timing θm1 to the advance side or the retard side. Specifically, when the torque deviation amount ΔTr is positive, that is, when the actual torque Tra is smaller than the reference torque Tr1, the ignition timing is changed to the advance side by a predetermined amount Δθ1. On the other hand, when the torque deviation amount ΔTr is negative, that is, when the actual torque Tra is larger than the reference torque Tr1, the ignition timing is changed to the retard side by a predetermined amount Δθ2. However, Δθ1 and Δθ2 may be the same value or different values. When changing the ignition timing, the intake air amount is kept constant, and the actual torque Tra is adjusted to the target torque only by changing the ignition timing.

  Note that when the reference ratio of the fuel is the minimum mixture ratio of the inert gas, the actual torque Tra is always smaller than the reference torque Tr1. Therefore, when changing the ignition timing to eliminate the torque change, the ignition timing is always changed to the advance side.

  In the subsequent step S108, it is determined whether or not the actual torque Tra matches the target torque as the ignition timing is changed. Specifically, it is determined whether or not the actual torque Tra has reached the maximum torque when the gaseous fuel currently supplied to the first injection valve 21 is used. Whether or not the maximum torque has been reached is determined based on, for example, a differential value of the actual torque Tra. If the deviation between the actual torque Tra and the target torque is still large, a negative determination is made in step S108, the process returns to step S107, and the ignition timing is advanced or retarded repeatedly.

  If it is determined that the actual torque Tra matches the reference torque Tr1, the process proceeds to step S109, where the ignition timing change is stopped and the ignition timing change amount required to make the actual torque Tra match the target torque is calculated. This is stored as the ignition advance amount Δθig. The ignition advance amount Δθig is set to a larger value as the change amount of the ignition timing to the advance side is larger, and is set to a smaller value as the change amount to the retard side is larger. Steps S107 to S109 correspond to ignition control means.

  In step S110, the combustion speed Vb is estimated based on the ignition advance amount Δθig. Here, an estimated value of the combustion speed Vb is calculated by reading a combustion speed corresponding to the current ignition advance amount Δθig using a predetermined combustion speed estimation map. FIG. 4 shows an example of the combustion speed estimation map. According to this map, the larger the ignition advance amount Δθig, the lower the speed value is set as the estimated value of the combustion speed Vb.

  In the subsequent step S111, the mixing ratio (inert gas ratio αi) of the inert gas in the gaseous fuel is calculated based on the estimated value of the combustion speed Vb. Here, the inert gas ratio αi is calculated by reading the inert gas ratio corresponding to the estimated value of the combustion speed Vb using a predetermined inert gas calculation map. FIG. 5 shows an example of an inert gas calculation map. According to this map, a larger value is set as the inert gas ratio αi as the combustion speed Vb is lower.

  Thereafter, in step S112, the calculated inert gas ratio αi is stored in the backup memory as a learning value, and this routine is terminated. Note that the processing of steps S110 to S112 corresponds to learning execution means. The inert gas ratio αi stored as the learned value of the fuel composition learning is used for fuel injection amount correction and ignition timing correction (see FIG. 7).

  Next, an abnormality diagnosis process for the intake pressure sensor 82, which is a process for determining whether or not the condition (2) is satisfied, will be described. In the present embodiment, the abnormality diagnosis process for the intake pressure sensor 82 is performed during fuel cut. Gaseous fuel has a larger volume per unit mass than liquid fuel, and when the gaseous fuel is injected, the value of the intake pressure sensor 82 changes due to the influence of the injected fuel. This is because it is considered impossible.

  The abnormality diagnosis process of the intake pressure sensor 82 will be described with reference to FIG. This process is executed by the control unit 80 at predetermined intervals. In FIG. 6, in step S201, it is determined whether or not the engine 10 is under fuel cut. If the fuel is not being cut, this routine is terminated. If the fuel is being cut, the process proceeds to step S202. In step S202, the throttle opening is changed to a predetermined diagnosis opening. The predetermined diagnosis opening is set to a position slightly on the valve opening side of the fully closed position of the throttle valve 15.

  In the subsequent step S203, the detected value of the intake pressure sensor 82 is acquired in a state where the throttle opening is set to a predetermined diagnostic opening, and in step S204, the deviation between the acquired sensor detected value and a predetermined normal reference value is obtained. It is determined whether it is within a predetermined range. If the deviation between the sensor detection value and the normal reference value is within a predetermined range, the process proceeds to step S205, and it is determined that the intake pressure sensor 82 is normal. On the other hand, if the deviation between the sensor detection value and the normal reference value is out of the predetermined range, the process proceeds to step S206, and it is determined that the intake pressure sensor 82 is abnormal. In this case, in FIG. 2, it is determined that the learning execution condition is not satisfied, and the execution of the fuel composition learning is prohibited.

  In this system, the fuel injection amount and the ignition timing are corrected based on the inert gas ratio αi as the learning value of the fuel composition learning. Hereinafter, the fuel injection amount correction and the ignition timing correction will be described with reference to the flowchart of FIG. This process is executed by the control unit 80 at predetermined intervals.

  In FIG. 7, in step S301, it is determined whether or not gaseous fuel has been replenished in the gas tank 42 (replenishment determining means). When the gaseous fuel is not replenished, correction based on the learned value of the fuel composition is not necessary, so in the present embodiment, this routine is terminated as it is. On the other hand, if it is determined that gas fuel is replenished, the process proceeds to step S302, and it is determined whether or not fuel composition learning has been performed after the current engine start (learning determination means). If the fuel composition learning has not been completed yet, this routine is once ended. If the learning has been completed, the process proceeds to step S303.

  In step S303, the inert gas ratio αi stored as the learning value is read from the backup memory, and in step S304, the fuel injection amount is corrected based on the read inert gas ratio αi. Specifically, a correction based on the inert gas ratio αi is applied as a fuel composition correction to the basic injection amount calculated based on the engine operating state. At this time, the fuel injection amount is corrected to the increase side as the inert gas ratio αi increases. In step S305, a correction based on the inert gas ratio αi is performed as a fuel composition correction on the base value (basic ignition timing) of the ignition timing calculated based on the engine operating state. At this time, the ignition timing is corrected to the advance side as the inert gas ratio αi increases. Thereafter, this routine is terminated.

  According to the embodiment described in detail above, the following excellent effects can be obtained.

  Focusing on the fact that the relationship between the actual torque generated in the engine 10 and the ignition timing differs depending on the fuel composition, when it is detected that a torque change accompanying the replenishment of gaseous fuel into the gas tank 42 has occurred, the torque The ignition timing is changed until the actual torque matches the predetermined target torque on the side of canceling the change, and the fuel composition learning is executed based on the amount of change in the ignition timing at that time. In this configuration, since the fuel composition is learned based on the change amount of the ignition timing, the fuel composition learning can be performed without being affected by changes in the fuel system. Accordingly, for example, the situation is more susceptible to changes in the fuel system than when the fuel composition is learned based on the air-fuel ratio correction amount in the air-fuel ratio feedback control (for example, aging of the first injection valve 21, purging, etc.). Can improve the learning accuracy.

  An ignition advance amount Δθig is calculated as an ignition timing change amount for eliminating the torque change, and fuel composition learning is executed based on the calculated ignition advance amount Δθig. Since there is a correlation between the ignition advance amount Δθig and the combustion speed Vb, and there is a correlation between the combustion speed Vb and the inert gas ratio αi (see FIGS. 4 and 5), according to the ignition advance amount Δθig, Currently, the mixing ratio of the inert gas in the gaseous fuel supplied to the first injection valve 21 can be grasped. In addition, since the mixing ratio of the inert gas in the fuel varies greatly depending on the region, by providing the mixing ratio of the inert gas as a learning value, the steady state of the fuel injection amount and the ignition timing due to the change in the composition of the gaseous fuel. Can be eliminated.

  The fuel composition learning is executed on condition that the engine load fluctuation is within a predetermined load fluctuation range and the engine rotation fluctuation is within a predetermined rotation fluctuation range. In this case, when learning the fuel composition, it is difficult to be affected by a torque change caused by a change in the fuel injection amount or the intake air amount, and erroneous learning can be avoided.

  Specifically, after the engine is started after the gas fuel is replenished in the gas tank 42, the ignition timing is set to the optimum ignition timing (reference optimum timing θm1) when the composition of the gaseous fuel is the reference composition. When a torque change sometimes occurs, the ignition advance amount Δθig is calculated as the ignition timing change amount necessary to eliminate the torque change, and the fuel composition is learned based on the calculated ignition advance amount Δθig. It was set as the structure to do. According to such a configuration, the fuel composition can be learned by actively moving the ignition timing after the operation of the engine 10 is started. In addition, the time required for learning can be shortened.

  According to the above configuration, the learning value of the fuel composition can be calculated in consideration of the difference from the case where the gaseous fuel has the reference composition. Further, since the correction reference is constant, the learning value can be calculated based on the constant reference regardless of the composition of the gaseous fuel, and the learning accuracy can be stabilized. In particular, the optimum ignition timing, which is the ignition timing at which the torque is maximum, has a correlation with the combustion speed, and the combustion speed tends to be faster as the optimum ignition timing is on the retard side. Therefore, by using the optimum ignition timing as a reference, it is possible to accurately grasp the change in the combustion speed due to the fuel composition deviation. Thereby, the inert gas ratio αi based on the ignition advance amount Δθig can be calculated with high accuracy.

(Second Embodiment)
Next, a second embodiment will be described. In the first embodiment, the fuel composition learning is performed by stopping the torque feedback and actively changing the ignition timing for canceling the torque change accompanying the fuel replenishment. On the other hand, in the present embodiment, the fuel composition learning is performed by continuing the torque feedback during the fuel composition learning and changing the ignition timing for canceling the torque change by controlling the event by the torque feedback. Hereinafter, the second embodiment will be described focusing on differences from the first embodiment.

  FIG. 8 is a flowchart showing a processing procedure of the fuel composition learning process of the present embodiment. This process is performed by the control unit 80 every time the vehicle is driven, for example. In the description of FIG. 8, the same processing as in FIG. 3 is denoted by the step number of FIG. 3, and the description thereof is omitted.

  In FIG. 8, in steps S401 to S402, the same processing as steps S101 to S102 in FIG. 3 is executed. In a succeeding step S403, it is determined whether or not a learning execution condition is satisfied. Here, in addition to the conditions (1) and (2), the learning execution conditions further include a condition that (3) the torque feedback control is started. And it progresses to step S404 on condition that all the conditions including these (1)-(3) are materialized.

  In step S404, a deviation of the actual torque Tra from the required torque Trq is calculated, and this is set as a torque deviation amount ΔTr (ΔTr = Trq−Tra). Note that when calculating the torque deviation amount ΔTr, for example, the actual torque Tra immediately after the start of the torque feedback control is used. In step S405, it is determined whether or not the torque deviation amount ΔTr is out of the predetermined allowable range. If it is within the allowable range, this routine is ended as it is. On the other hand, if the torque deviation amount ΔTr is out of the allowable range, the process proceeds to step S406. Note that the processing in steps S402 to S405 corresponds to change detection means.

  In step S406, the ignition timing is advanced or retarded so that the actual torque Tra matches the required torque Trq with the air amount kept constant. For example, the gas fuel having the fuel composition of the reference composition was filled in the gas tank 42 at the time of the previous engine operation, whereas the gas fuel having a higher mixing ratio of the inert gas than the reference composition at the time of the current engine operation. Let us consider a case where the gas tank 42 is filled. In this case, immediately after the current engine start, the actual torque Tra becomes lower than the required torque Trq (for example, point B in FIG. 2). At this time, ignition timing advance angle correction is performed by performing torque feedback with a constant intake air amount (for example, point C in FIG. 2). In the present embodiment, a learning value for fuel composition learning is calculated based on this ignition correction amount.

  Specifically, in step S407, it is determined whether or not the actual torque Tra matches the target torque (required torque Trq in the present embodiment) as the ignition timing is changed. If a negative determination is made in step S407, the process returns to step S406 to continue torque feedback with a constant air amount. If it is determined that the actual torque Tra matches the required torque Trq, the process proceeds to step S408, and the ignition timing correction amount required to make the actual torque Tra match the required torque Trq is calculated, and this is corrected. Stored as a quantity Δθk. The ignition correction amount Δθk is set to a larger value as the change amount of the ignition timing to the advance side is larger, and is set to a smaller value as the change amount to the retard side is larger. Steps S406 to S408 correspond to ignition control means.

  Thereafter, in step S409, the combustion speed Vb is estimated based on the ignition correction amount Δθk. In steps S410 to S411, processing similar to that in steps S111 to S112 in FIG. 3 is executed, and this routine is terminated. The inert gas ratio αi stored as the learned value in the present embodiment is also used for fuel injection amount correction and ignition timing correction, as in the first embodiment.

  In the second embodiment described in detail above, the gaseous fuel into the gas tank 42 is based on the deviation between the actual torque Tra and the required torque Trq after the engine operation is started after the gaseous fuel is replenished into the gas tank 42. The fuel composition learning is executed on the basis of the ignition timing correction amount Δθk for eliminating the deviation in a state where the air amount is constant while detecting that the torque change caused by the replenishment of the fuel has occurred. According to such a configuration, the ignition timing can be moved according to the torque feedback. In addition, since torque feedback control is continued, torque fluctuation is unlikely to occur, and the fuel composition can be learned while continuing stable combustion.

(Other embodiments)
The present invention is not limited to the above embodiment, and may be implemented as follows, for example.

  In the first embodiment, the combustion speed Vb is estimated using the combustion speed estimation map of FIG. 4 based on the ignition advance amount Δθig, and the inert gas calculation of FIG. 5 is calculated based on the estimated combustion speed Vb. The inert gas ratio αi is calculated using a map. By changing this, a map in which the relationship between the ignition advance amount Δθig and the inert gas ratio αi is set in advance is used, and the inert gas ratio αi is calculated based on the current ignition advance amount Δθig using this map. It is good. Also in the second embodiment, a map in which the relationship between the ignition correction amount Δθk and the inert gas ratio αi is set in advance is used, and the inert gas ratio αi is calculated based on the current ignition correction amount Δθk using this map. It is good also as composition to do.

  In the first embodiment, the maximum torque generated when the gaseous fuel (fuel in the gas tank 42) supplied to the first injection valve 21 burns is set as the target torque. However, the present invention is not limited to this. For example, a torque lower by a predetermined amount than the maximum torque may be set as the target torque.

  In the above-described embodiment, it may be configured to detect a change in torque caused by replenishment of gaseous fuel into the gas tank 42 by detecting a change in engine rotation speed. For example, during idle operation, gaseous fuel into the gas tank 42 is detected by detecting that the deviation of the actual value of the engine speed (actual speed) from the target idle speed is outside a predetermined allowable range. It is configured to detect that a torque change accompanying the replenishment of. In this case, an inert gas ratio αi is calculated as a learned value of the fuel composition based on the ignition advance amount for making the actual rotational speed coincide with the target idle rotational speed.

  In the first embodiment, the fuel composition learning is performed with the gas fuel having the reference composition as a correction reference. However, the correction reference is not limited to this, for example, the ignition timing at the time of engine operation before this time Fuel composition learning may be performed based on the relationship between torque, fuel, and fuel composition (learned value).

  In the above embodiment, (1) when the engine load fluctuation is within a predetermined load fluctuation range and the engine rotation speed is within the predetermined rotation fluctuation range, the engine steady state during the idling operation state and the vehicle running The configuration includes a state. In other words, the fuel composition learning is performed on the condition that the engine is in the idle operation state or the engine is in a steady state while the vehicle is running. On the other hand, only the idling state may be included in the learning execution condition, and the fuel composition learning may not be performed when the engine is in a steady state while the vehicle is running. On the contrary, only the engine steady state while the vehicle is running may be included in the learning execution condition, and the fuel composition learning may not be performed in the idle operation state. In particular, during idle operation, a state in which the load fluctuation and rotation fluctuation of the engine 10 are small can be secured for a relatively long time, which is suitable for executing fuel composition learning.

  In the above embodiment, the mixing ratio of the inert gas in the gaseous fuel is calculated as the learning value of the fuel composition. However, any value that correlates with the mixing ratio of the inert gas may be used. For example, the density of the gaseous fuel May be calculated as a learning value. This is because if the mixing ratio of the inert gas is different, the density of the fuel changes depending on the mixing ratio.

  In the first embodiment, the case where the required torque is calculated based on the engine operating state and applied to a system that operates the engine by torque feedback control to match the actual torque with the required torque has been described. The present invention may be applied to an engine operation system that does not depend on the present invention.

  In the above-described embodiment, the case where it is applied to a bi-fuel type vehicle-mounted engine that uses gaseous fuel and liquid fuel as engine combustion fuel has been described, but only for gas that uses only gaseous fuel as fuel for engine fuel You may apply to the vehicle-mounted engine.

  In the above embodiment, the gaseous fuel is CNG fuel, but other gaseous fuels in a gaseous state can be used in the standard state, for example, a fuel mainly composed of methane, ethane, propane, butane, hydrogen, dimethyl ether, etc. It is good also as a structure to use. Further, liquid fuel is not limited to gasoline fuel. For example, you may apply this invention to the system which mounts the supply system of gaseous fuel with respect to the diesel engine which uses light oil as a fuel for combustion.

  DESCRIPTION OF SYMBOLS 10 ... Engine (internal combustion engine), 20 ... Spark plug, 20a ... Ignition device, 21 ... 1st injection valve (gas injection means), 22 ... 2nd injection valve, 40 ... Gas fuel supply part, 41 ... Gas piping, 42 ... Gas tank (fuel tank), 43 ... Regulator, 70 ... Liquid fuel supply unit, 80 ... Control unit (change detection means, ignition control means, learning execution means, state determination means, replenishment determination means, learning determination means, injection amount calculation Means, ignition calculation means).

Claims (7)

Applied to an internal combustion engine (10) comprising a fuel tank (42) for storing gaseous fuel in a high pressure state, and gas injection means (21) for injecting gaseous fuel supplied from the fuel tank through a fuel passage (41),
A change detecting means for detecting that a torque change associated with the replenishment of the gaseous fuel into the fuel tank has occurred with respect to the actual torque generated in the internal combustion engine;
Ignition control means for changing the ignition timing of the internal combustion engine until the actual torque reaches a predetermined target torque for eliminating the torque change when the change detection means detects that the torque change has occurred. When,
Learning execution means for performing composition learning of the gaseous fuel supplied to the gas injection means based on the amount of change in the ignition timing by the ignition control means;
A control device for an internal combustion engine, comprising:   2. The internal combustion engine according to claim 1, wherein the learning execution unit calculates a mixing ratio of an inert gas in the gaseous fuel or a correlation value thereof as the composition learning based on a change amount of an ignition timing by the ignition control unit. Control device. A state determination unit that determines whether or not the load fluctuation of the internal combustion engine is within a predetermined load fluctuation range and the rotation speed of the internal combustion engine is within a predetermined rotation fluctuation range;
The control device for an internal combustion engine according to claim 1 or 2, wherein the learning execution means executes the composition learning on condition that the state determination means determines that the learning operation state is established. Replenishment determination means for determining whether or not the gaseous fuel has been replenished into the fuel tank;
Learning determination means for determining whether or not the composition learning by the learning execution means is completed;
An injection amount calculating means for calculating a basic injection amount that is a base value of a fuel amount injected by the gas injection means based on an operating state of the internal combustion engine;
With
When it is determined by the replenishment determination means that the gaseous fuel has been replenished, the basic injection amount is corrected based on the learning result of the learning execution means after the learning determination means determines that the composition learning has been completed. The control device for an internal combustion engine according to any one of claims 1 to 3. Replenishment determination means for determining whether or not the gaseous fuel has been replenished into the fuel tank;
Learning determination means for determining whether or not the composition learning by the learning execution means is completed;
Ignition calculation means for calculating a basic ignition timing that is a base value of the ignition timing of the internal combustion engine based on the operating state of the internal combustion engine;
With
When it is determined by the replenishment determination means that the gaseous fuel has been replenished, the basic ignition timing is corrected based on the learning result of the learning execution means after the learning determination means determines that the composition learning has been completed. The control device for an internal combustion engine according to any one of claims 1 to 4. A standard composition is defined as the composition of the gaseous fuel,
The change detecting means is configured to set the ignition timing of the internal combustion engine to the optimum ignition timing when the composition of the gaseous fuel is the reference composition after the gaseous fuel is replenished in the fuel tank. Based on the torque change, it detects that a torque change accompanying the replenishment of the gaseous fuel into the fuel tank has occurred,
6. The ignition control unit according to claim 1, wherein the target torque is set to a maximum torque generated when the gaseous fuel supplied to the gas injection unit burns, and the ignition timing is changed until the maximum torque is reached. The control apparatus for an internal combustion engine according to the item. Torque feedback control is performed so that the actual torque matches the required torque calculated based on the operating state of the internal combustion engine,
The change detecting means is configured to change a torque accompanying replenishment of the gaseous fuel into the fuel tank based on a deviation between the actual torque and the required torque after the gaseous fuel is replenished into the fuel tank. Detected that
6. The ignition control unit according to claim 1, wherein the ignition control unit changes the ignition timing so that the actual torque becomes the required torque when the target torque is the required torque and the air amount is constant. Control device for internal combustion engine.

Images