JP5015877B2 - Engine control device - Google Patents

Description

  The present invention relates to an engine control device that controls a fuel injection amount in accordance with fuel properties.

  2. Description of the Related Art Conventionally, a technique is known in which fuel properties are determined from the rotational speed behavior at the time of engine start, and fuel injection control is suitably performed based on the determination result of the fuel properties. The technology for determining the fuel properties from the behavior of the engine speed and reflecting it in the fuel injection amount control is as follows: “Because heavy fuel has low volatility, the combustion state becomes lean and the speed behavior becomes unstable. In addition, light fuel is highly volatile, so the combustion state becomes rich and the rotational speed behavior is stable. ”Heavy fuel is used when the rotational speed behavior is unstable. It is determined that the fuel injection amount is increased.

  Further, in a general gasoline engine, fuel injection called intake port injection is performed, and fuel is injected from an injector arranged in the intake port toward an intake valve arranged in the engine combustion chamber. In such intake port injection, a part of the injected fuel adheres to the intake valve and the intake port and is then vaporized and sucked into the cylinder. The fuel adhering to the intake port or the intake valve without being vaporized is called port wet.

  At the time of cold start, the injected fuel cannot be sufficiently vaporized because the port temperature is low. In order to compensate for the shortage of fuel vaporization, it is necessary to inject more fuel than after the engine is warmed up, and port wetness is also increased. Since the port wet contributes to combustion in the next cycle, for example, in the case of a four-cylinder engine, it is considered that it changes stepwise every four ignitions (or every four injections) in the initial stage of engine startup.

A technique for correcting the fuel injection amount in consideration of this port wet is disclosed in Patent Document 1, for example. In the technique of Patent Document 1, the ratio of fuel injected from the fuel injection valve becomes port wet (attachment rate) and the ratio of the fuel attached to the wall surface remaining as port wet without being vaporized in the course of the intake stroke ( The residual ratio is used to create a model representing the behavior of the fuel after injection, and the desired fuel injection amount is calculated with high accuracy.
Japanese Patent Laying-Open No. 2005-9467

  Fuel properties in the market are widely distributed from high volatility (light fuel properties) to low volatility (heavy fuel properties). For this reason, at the initial stage of engine starting, that is, at the stage where the influence of port wet changes greatly, it is necessary to appropriately determine the fuel properties and to make the fuel injection amount appropriate. In particular, in the case of fuel with low volatility, the air-fuel mixture flowing into the cylinder becomes lean and the rotational speed behavior tends to become unstable.

  However, when determining the fuel properties, the rotational speed behavior at the time of starting the engine is affected by effects other than the fuel properties, so there are various situations that deviate from the above-mentioned idea. For example, a situation where the combustible air is filled upstream of the intake system due to oil-tight leakage of the injector, the air-fuel mixture supplied to the cylinder becomes over-rich, and the rotational speed behavior becomes unstable. In situations where the fuel ratio varies and the rotational speed behavior becomes unstable, in high altitudes, the air density supplied to the engine is small, so that sufficient power cannot be obtained and the rotational speed behavior becomes unstable. This is because, for example, the engine friction varies with time, so that the rotational speed behavior is different.

  Therefore, it is difficult to accurately determine the fuel property from the rotational speed behavior, and the fuel property may be erroneously determined. If a misjudgment of the fuel property occurs, the controllability of the fuel injection amount deteriorates and the exhaust emission deteriorates. On the other hand, after the engine is started, if the negative pressure increases due to the increase in the rotational speed, fuel vaporization is promoted, and the necessity of increasing the fuel injection amount gradually disappears. Therefore, if the port wet is uniquely considered, the amount of fuel increase becomes excessive, and the exhaust emission may be deteriorated.

  The present invention has been made in view of the above circumstances, and focuses on the action of port wet to increase the accuracy of determination of fuel properties and to achieve both improvement in starting performance and reduction in exhaust emission. The purpose is to provide.

In order to achieve the above object, an engine control apparatus according to the present invention includes a port wet detection unit that detects port wet that changes according to the fuel injection amount and the number of injections of fuel injected into the intake system, and the port wet detection unit. A fuel property determination unit that determines the fuel property of the used fuel from the combustion state in association with the port wet detected by the fuel, and a port that is detected by the port wet detection unit from the fuel property determined by the fuel property determination unit A fuel injection amount correction unit that corrects the fuel injection amount in association with wet, and the port wet detection unit defines a port wet cycle in which the number of ignitions or the number of injections with the same port wet is set as one processing unit The port wet is detected for each processing unit of the port wet cycle, and the fuel is The state determination unit defines a combustion state detection cycle in which a period for detecting the combustion state of one processing unit of the port wet cycle is further set as one processing unit, and each processing unit of the port wet cycle and the combustion state detection cycle are defined. Each fuel processing unit is associated with each other, and the fuel property is determined for each processing unit of the combustion state detection cycle .

  According to the present invention, it is possible to increase the accuracy of determination of fuel properties by paying attention to the action of port wet, and it is possible to achieve both improvement in engine starting performance and reduction in exhaust emission.

  Embodiments of the present invention will be described below with reference to the drawings. 1 to 20 relate to an embodiment of the present invention, FIG. 1 is a configuration diagram of an engine control system, FIG. 2 is a flowchart showing main processing, FIG. 3 is a flowchart of initial processing, and FIG. 5 is a flowchart of the combustion state determination result acquisition process, FIG. 6 is a flowchart of the fuel property determination execution condition establishment determination process, FIG. 7 is a port wet cycle calculation process flowchart, and FIG. 8 is a combustion state determination value calculation process. 9 is a flowchart of the combustion state determination count calculation process, FIG. 10 is a flowchart of the fuel property determination process, FIG. 11 is a flowchart of the injection amount magnification calculation process, FIG. 12 is a flowchart of the learning value writing process, and FIG. FIG. 14 is a flowchart of calculation processing, and FIG. 14 shows heavy fuel and light fuel based on the fuel injection amount of intermediate fuel. 15 is an explanatory diagram showing the injection amount magnification of the fuel, FIG. 15 is an explanatory diagram showing the relationship between the output timing of the combustion state determination result and each processing timing, and FIG. 16 is a heavy fuel fuel determination pattern from heavy fuel to intermediate fuel FIG. 17 shows the behavior of the injection amount magnification attenuation coefficient when the heavy fuel determination is repeated in the fuel property determination pattern from the heavy fuel to the intermediate fuel when the fuel determination is not made. FIG. 18 is an explanatory diagram showing the behavior of the injection amount magnification attenuation coefficient when the subtraction amount is reduced outside the fuel property learning value update period in the fuel property determination pattern from heavy fuel to intermediate fuel, and FIG. FIG. 20 is an explanatory view showing the behavior of the injection amount magnification attenuation coefficient when subtracting to near 0 within the fuel property learning value update period in the fuel property determination pattern from the quality fuel to the intermediate fuel. Is an explanatory view showing the relationship between the cycle and the combustion state detection cycle.

  In FIG. 1, reference numeral 1 denotes an engine, and a cylinder block 2 of the engine 1 is provided with a plurality of cylinders 3 (only one is shown in the figure). In each cylinder 3, a piston 4 is disposed so as to be able to reciprocate, and a combustion chamber 6 is formed between the top surface of the piston 4 and the cylinder head 5.

  The cylinder head 5 is provided with an intake port 7 and an exhaust port 8, one end of these ports 7 and 8 is connected to the combustion chamber 6, and the other end is connected to an intake passage 9 and an exhaust passage 10. An intake valve 11 and an exhaust valve 12 are provided at each boundary surface between the combustion chamber 6 and the ports 7 and 8, respectively. These intake valve 11 and exhaust valve 12 are provided in the cylinder head 5. Reciprocatingly driven by the drive mechanism, the combustion chamber 6 is opened and closed.

  An air filter (not shown), a throttle valve 13, a collector 14, and an intake manifold 15 are provided in the intake passage 9 from the upstream side, and the intake manifold 15 is connected to the intake port 7. On the other hand, an exhaust manifold 16 connected to the exhaust port 8 and a catalytic converter (not shown) are provided in the exhaust passage 10 from the upstream side. In the catalytic converter, Nox (nitrogen oxide) generated by the oxidation of incomplete combustion components HC (hydrocarbon) and CO (carbon monoxide) and the reaction of nitrogen in the air with unburned oxygen. A three-way catalyst that simultaneously promotes reduction of the catalyst is housed.

  The above engine 1 is provided with various actuators. That is, a throttle actuator 17 is connected to the throttle valve 13, and the throttle valve 17 is opened and closed by the throttle actuator 17. In addition, an injector 18 that injects fuel toward the intake valve 11 and the intake port 7 is attached to the intake manifold 15.

  Further, the cylinder head 5 is provided with an ignition plug 19 for igniting the air-fuel mixture taken into the combustion chamber, and an ignition coil 20 connected to the ignition plug 19. The ignition coil 20 is disconnected from the primary current during ignition, and the secondary current is supplied to the spark plug 19.

  Various sensors are attached to the engine 1. That is, a water temperature sensor 21 for detecting the temperature of the cooling water of the engine 1 is attached to the cylinder block 2, and an air flow meter 22 for detecting the amount of intake air is detected in the intake passage 9 and the temperature of the intake air is detected. An intake air temperature sensor 23 is attached.

  Further, a throttle opening sensor 24 for detecting the rotation angle of the shaft and an idle switch (not shown) for detecting the fully closed state of the throttle valve 13 are disposed in the vicinity of the throttle valve 13. Has been. The collector 14 is provided with an intake pressure sensor 25 for detecting the internal pressure.

  The exhaust manifold 16 also outputs an air-fuel ratio (A / F) sensor that outputs a wide-range and linear air-fuel ratio signal in proportion to the oxygen concentration in the exhaust gas (or the concentration of carbon monoxide, which is unburned gas). 26 is attached. Further, a crank angle sensor 27 for detecting the engine speed is disposed in the vicinity of the crankshaft of the engine 1.

  Next, the electronic control system of the engine 1 will be described. The electronic control system according to the present invention is represented by a combustion state detection device 30 and an engine control device (ECU) 50. Hereinafter, these apparatuses will be described.

  The combustion state detection device 30 is a combustion state determination that determines a combustion state based on an ion current detection circuit 31 connected to the secondary winding side of the ignition coil 20 and an ion current behavior that is an output of the ion current detection circuit 31. And a portion 32. The combustion state detection device 30 and the ECU 50 are communicably connected to each other, and the combustion state determination result in the combustion state determination unit 32 is transmitted to the ECU 50.

  The ECU 50 is mainly configured of a microcomputer including a CPU, ROM, RAM, backup RAM, and the like. The ECU 50 inputs signals from the various sensors, switches, and combustion state detection device 30 described above, and executes engine control in accordance with a program in the ROM. The backup RAM is a memory that retains stored contents even when the ignition switch is turned off by power supply from a battery, and various learning values are stored and retained.

  Here, typical control of the engine control by the ECU 50, that is, ignition timing control, intake air amount control, and fuel injection control will be described.

  In the ignition timing control, the ECU 50 calculates the optimum ignition timing according to the program in the ROM with reference to the signals of various sensors and switches, and converts the ignition coil energization period for realizing the ignition timing into the crank angle timing. Then, an energization signal is sent to the primary coil of the ignition coil 20 based on the crank angle timing. At the time of cold start, the ignition timing is controlled to the retarded side for early activation of the catalyst.

  In the intake air amount control, the ECU 50 refers to the target idle speed determined from various sensors, switch signals, and engine coolant temperature, and the intake air that can maintain the target speed according to the program in the ROM. Calculate the quantity. Then, the throttle actuator 17 is driven to the throttle opening that realizes the intake air amount. At the time of cold start, a fast idle speed higher than the idle speed after warm-up is set as the target speed for early activation of the catalyst.

  In the fuel injection control, the ECU 50 calculates the fuel injection amount according to the program in the ROM with reference to the determination result of the fuel property determined based on the combustion state from the combustion state detection device 30 and the signals of various sensors and switches. Then, the injector driving period for realizing the fuel injection amount is converted into the crank angle timing. Then, a drive signal is sent to the injector 18 based on the crank angle timing. The fuel injection control includes air-fuel ratio feedback control that uses the air-fuel ratio signal of the A / F sensor 26 to feedback-control the fuel injection amount so that the air-fuel ratio becomes a predetermined value.

  The function of the ECU 50 in the fuel injection control according to the present invention is based on the result of combustion state determination from the port wet detection unit 51 and the combustion state detection device 30 that detect fuel (port wet) adhering to the intake system without being vaporized. The fuel property determination unit 52 that determines the property, and the fuel injection amount correction unit 53 that corrects the fuel injection amount in association with the port wet from the fuel property determined by the fuel property determination unit 52 can be represented. The fuel property determined by the fuel property determination unit 52 is learned, and the learned value is stored and retained in the backup RAM.

  The functions of the port wet detection unit 51, the fuel property determination unit 52, and the fuel injection amount correction unit 53 are specifically realized by the program processing shown in the flowcharts of FIGS. Next, the function of each part will be described in detail according to the flowcharts of FIGS.

  FIG. 2 shows a main process showing the overall control according to the present invention. Steps S2 and S5 are for the port wet detection unit 51, steps S3 and S6 to S8 are for the fuel property determination unit 52, and steps S9 to S9 are steps. S 11 corresponds to the fuel injection amount correction unit 53. Details of each step in the main process are shown in the flowcharts of FIGS.

  The flow of the main process is the initial process in step S1, the number of ignitions is calculated in step S2, and the combustion state determination result is acquired from the combustion state detection device 30 in step S3. Next, in step S4, it is checked whether or not a condition for determining whether the fuel property is heavy fuel or light fuel is satisfied. If the condition is not satisfied, the process proceeds to an injection amount calculation process in step S11. If the condition is satisfied, step S5 is performed. The process proceeds to the port wet cycle calculation process.

  In the step of determining whether the condition is satisfied in the flowchart, T indicates that the condition is satisfied (true), and F indicates that the condition is not satisfied (false).

  As will be described later, in the port wet cycle, the number of ignitions or the number of injections with the same port wet is set as one processing unit, and each processing of steps S6 to S9 (for each processing unit of the port wet cycle ( The calculation of the combustion state determination value, the calculation of the combustion state determination count, the fuel property determination, the calculation of the injection amount magnification) is executed, the fuel property learning value in step S10 is written to the backup RAM, and the fuel injection amount in step S11 is calculated. After the calculation process, the main process ends. Hereinafter, the contents of each step of the main process will be described in detail.

[Initial processing]
In the main process of FIG. 2, the initial process is performed in the first step S1. This initial process is a process for obtaining an appropriate fuel injection amount commensurate with the fuel used from the beginning of the engine start. By avoiding excessively reducing the fuel injection amount, deterioration of the start performance is avoided, and exhaust gas performance is avoided. To prevent the deterioration.

  That is, in general, as the fuel injection amount at the initial stage of engine start, the injection amount at the time of using heavy fuel with the largest injection amount is used in order to ensure the starting performance. That is, exhaust gas performance deteriorates when using intermediate fuel or light fuel that requires less injection than heavy fuel. Therefore, in the initial process of step S1, the learning value of the fuel property is read before starting the engine, and the fuel injection amount is appropriate for the fuel used from the beginning of the engine starting, thereby preventing the exhaust gas performance from deteriorating.

  However, if there is no experience of rewriting the learned value, select the heavy fuel with the largest injection amount, and if the learned value is light fuel, it may be affected by the injector oil tight leak at the previous start. (There is a possibility that rich air-fuel mixture is supplied into the cylinder until the air upstream of the intake system is filled with combustible air due to the injector oil tight leak and the air upstream of the intake system is scavenged.) By avoiding a small amount of light fuel and selecting an intermediate fuel that is an average injection amount, deterioration of the starting performance is prevented.

  Specifically, the initial process is executed according to the procedure shown in FIG. First, when the ignition switch is turned on, the engine starting water temperature (the engine water temperature when the ignition switch is turned on) is a lower limit temperature kEXETWL (for example, kEXTETWL = −15 ° C.) and an upper limit temperature kEXETWH (for example, kEXTETWH = 30 ° C.). The condition within the set range is checked (step S1-1). In this condition, the volatility of the fuel with respect to the temperature varies depending on the fuel properties, so that the temperature range in which a significant difference in the fuel injection amount due to the fuel properties is reflected in the fuel property learning value by referring to the engine coolant temperature. It is a condition.

  As a result, when the condition is not satisfied in step S1-1, the injection amount magnification taupamp is set to 1.0 (step S1-6), and the initial process is terminated. The injection amount magnification tauamp is a magnification for increasing / decreasing the fuel injection amount according to the fuel properties with the fuel injection amount when using the intermediate fuel as a reference (intermediate fuel injection amount magnification tauamp = 1.0).

  On the other hand, when the condition is satisfied in step S1-1, appropriate values are set in various parameters (step S1-2), and the process proceeds to step S1-3. Details of each parameter will be described later, but the processing in step S1-2 is mainly processing such as initialization of a counter or flag, determination value, the number of experiences for the determination value, setting of an initial value to a coefficient, etc. Also, the port wet cycle wetcycle is set to 1. The port wet cycle wetcycle has the number of ignitions or the number of injections with the same degree of port wet as one processing unit, and is calculated in the process of step S5 described later.

  In step S1-3, a condition of rewriting experience in the backup RAM is checked against the learned value fuel_buram obtained by learning the fuel property determination value fuel, which is the fuel property determination result. In the present embodiment, fuel = 1 is a light fuel, fuel = 2 is an intermediate fuel, and fuel = 3 is a heavy fuel.

  When the condition of step S1-3 is satisfied (with rewriting experience), the fuel property determination result at the previous start is checked (step S1-4), and as a result, the fuel property determination result at the previous start is intermediate fuel or light In the case of fuel (fuel_buram ≦ 2), the intermediate fuel is designated as the fuel to be used (step S1-5). Then, the injection amount magnification 1.0 of the intermediate fuel is set to the injection amount magnification tauamp (step S1-6), and the initial process is terminated.

  On the other hand, if the condition of step S1-3 or step S1-4 is not satisfied, the heavy fuel is designated as the fuel to be used (step S1-7), and the wet cycle = 1 set in advance for the injection amount magnification taupamp. After the heavy fuel injection amount magnification tHEAVYAMP is set (step S1-8), the initial process is terminated. For example, as shown in FIG. 14, the injection amount magnification tHEAVYAMP preset for the heavy fuel is set for each port wet cycle, and in step S1-8, the injection amount magnification tHEAVYAMP = 1.18 for wetcycle = 1. Thus, tauamp = 1.18 is set.

[Ignition frequency calculation process]
Next, the process proceeds to step S2 of the main process in FIG. 2 to calculate the number of ignitions cntigexe. Specifically, as shown in FIG. 4, every time the ignition signal is output, the ignition number counter cnsize is incremented (step S2-1), and the process of step S2 is exited.

  As described above, in the present embodiment, the combustion state is determined from the ion current detection circuit 31 that detects the ion current corresponding to the amount of ions generated by the combustion of the air-fuel mixture in the combustion chamber, and the ion current behavior. The combustion state detection device 30 including the combustion state determination unit 32 communicates with the ECU 50 as a separate structure. The combustion state detection device 30 outputs the combustion state determination result to the ECU 50 in synchronization with the ignition timing (see FIG. 15).

[Acquisition processing of combustion state judgment results]
The next step S3 is a process of acquiring the combustion state determination result, and details are shown in FIG. This combustion state determination result is obtained by performing AD conversion of the combustion state determination result from the combustion state detection device 30 and storing it in the variable ionad for each calculation period of TDC-120 deg (step S3-1). The process exits and proceeds to step S4.

[Fuel property determination execution condition determination process]
Step S4 is a process for determining whether or not the “fuel property determination execution condition” is satisfied, and details are shown in FIG. This condition determination is performed every TDC-40 deg calculation cycle, and “the water temperature at engine start is within the range of the lower limit temperature kEXETWL (for example, kEXTETWL = −15 ° C.) and the upper limit temperature kEXETWH (for example, kEXETWH = 30 ° C.)” And “no experience of air-fuel ratio feedback execution” is checked (step S4-1).

  That is, the volatility of the fuel with respect to temperature varies depending on the fuel properties. Further, when the A / F sensor is activated and the air-fuel ratio feedback control is started ten seconds after the engine is started, the fuel injection amount correction by the air-fuel ratio feedback control and the fuel injection amount correction based on the fuel property by the fuel injection amount correction unit 53 are performed. Will overlap.

  Accordingly, in step S4-1, the engine coolant temperature is referred to, a temperature zone in which a significant difference in the fuel injection amount depending on the fuel property is set as the fuel property determination execution condition, and the fuel property determination execution period is set to the air-fuel ratio. Only before starting the feedback control, the fuel injection amount correction is avoided from becoming complicated, and the fuel injection amount correction by the air-fuel ratio feedback control is unified.

  In step S4-1, when the condition is satisfied, the process proceeds to the port wet cycle calculation process in step S5. When the condition is not satisfied, the process proceeds to the injection amount calculation process in step S11.

[Port wet calculation processing]
First, the port wet calculation process will be described. When the engine is cold-started, the temperature is low, so that more fuel needs to be injected and the port wetness also increases than after the engine is warmed up to compensate for the fuel vapor shortage. Since the port wet contributes to combustion in the next cycle, for example, in the case of a four-cylinder engine, it changes in stages at every four ignitions or every four injections in the initial stage of engine start.

  However, when a negative pressure is generated due to an increase in the engine speed and fuel vaporization is promoted, the necessity of increasing the fuel injection amount gradually disappears. For example, port wetness is grasped every 8 ignitions or every 8 injections. Will be able to. Therefore, a port wet cycle wetcycle is defined in which the number of ignitions or the number of injections with the same port wet is set as one processing unit, and each processing unit of this port wet cycle wetcycle is determined in time series as wetcycle = 1,2,3. ,... Are detected to detect port wet.

  In this case, in the present embodiment, since the combustion state determination result is transmitted from the combustion state detection device 30 to the ECU 50 by communication, each processing unit of wetcycle recognized on the ECU 50 side and the output from the combustion state detection system Must be associated with the combustion state determination result.

  Therefore, the combustion state detection cycle wetcycle is defined with the period for detecting the combustion state of one processing unit of the port wet cycle wetcycle as one processing unit. That is, as shown in FIG. 20, for each of wetcycle = 1, 2, 3,..., Each processing unit of the port wet cycle wetcycle and the port wet cycle are set as wetcycle = 1, 2, 3,. A period for detecting the combustion state of each processing unit of wetcycle is associated. However, wetcycle = 0 is a period for detecting a combustion state of ignition for a cylinder in which fuel injection is not performed during engine cranking, which is an initial operation of engine start.

  Here, when ignition for a cylinder that has not been subjected to fuel injection is defined as “invalid ignition” and ignition for a cylinder that has been subjected to fuel injection is defined as “effective ignition”, wetcycle = 0 is “the combustion state of invalid ignition” Is a determination result storage period ", and wecycle_out = 1,2,3... Becomes a" effective ignition combustion state determination result storage period for wetcycle = 1, 2, 3,. In a general internal combustion engine, invalid ignition is performed at least once at the time of engine cranking, which is the initial operation of engine start.

  Specifically, details of the port wet cycle calculation process are shown in FIG. This port wet cycle calculation process is executed every TDC-40 deg calculation cycle, and first, the ignition frequency counter continuge is checked (step S5-1). If the ignition is performed for the first time after the ignition switch is turned on, the synchronization counter cntsync is incremented (step S5-2), and the process proceeds to step S5-3. When the condition of step S5-1 is not satisfied, the process is exited.

  As shown in FIG. 15, in the present embodiment, the output timing of the combustion state determination result from the combustion state detection device 30 is equivalent to 540 deg CA after ignition, and therefore, according to the definition of the combustion state detection cycle wetcycle, the combustion state detection cycle It is necessary to synchronize the increment timing of wetcycle and the output timing of the combustion state determination result. Similarly, in accordance with the definition of the port wet cycle wetcycle, it is also necessary to synchronize the increment timing of the port wet cycle wetcycle and the injection timing.

  Therefore, in the present embodiment, a synchronization counter cntsync that uses the number of ignitions, an effective ignition number counter cntig, and an injection number counter ctinj are prepared, and synchronization is obtained by arbitrarily giving setting values for these counters. ing. By doing so, replacement with another system that detects the combustion state is facilitated without changing the control content.

  In step S5-3, the synchronization counter cntsync is compared with a set value kCNTIGSYNC (for example, kCNTIGSYNC = 5). If the counter value is not less than the set value kCNTIGSYNC, the effective ignition number counter cntig is incremented (step S5-4), The process proceeds to S5-5.

  In step S5-5, the condition “the combustion state detection cycle wetcycle is 0” or “the effective ignition number counter cntig is greater than the set value tIXCYCOUT” is checked. For example, the setting value tIXCYCOUT is set to tIXCYCOUT = 8 under the condition of wetcycle <40 and tIXCYCOUT = 4 under the condition of wetcycle ≧ 40.

  When the condition is satisfied, the effective ignition number counter cntig is set to 1 (step S5-6), the combustion state detection cycle wetcycle is incremented (step S5-7), and the process proceeds to step S5-8.

  On the other hand, when the conditions of step S5-3 and step S5-5 are not satisfied, the process proceeds to step S5-8. In step S5-8, the counter cntsync for synchronization is compared with a set value kCNTINJSYNC (for example, kCNTINJSYNC = 2), and if it is equal to or greater than the set value kCNTINJSYNC, the injection number counter ctinj is incremented (step S5-9), and then step S5 Move to -10.

  In step S5-10, the condition “the port wet cycle wetcycle is 1” or “the injection number counter cntinj is greater than the set value tIXCYC” is checked. For example, the set value tIXCYC is set to tIXCYC = 4 under the condition of wetcycle <40 and tIXCYC = 8 under the condition of wetcycle ≧ 40.

  When the condition is satisfied, the injection number counter cntinj is set to 1 (step S5-11), the port wet cycle wetcycle is incremented (step S5-12), the present port wet cycle calculation process is exited, and FIG. The process proceeds to the combustion state determination value calculation process in step S6. When the conditions in steps S5-8 and S5-10 are not satisfied, the process exits from the port wet cycle calculation process and proceeds to the combustion state determination value calculation process in step S6 of FIG.

[Calculation process of combustion state judgment value]
Next, “calculation of the combustion state determination value” in step S6 will be described. Details of the combustion state determination value calculation process in step S6 are shown in FIG. 8, and are executed every TDC-40 deg calculation cycle.

  In the combustion state determination value calculation process of FIG. 8, after the AD conversion value ionad of the combustion state determination result is replaced with a control desired numerical value and stored in the combustion state determination value ion for determining the combustion state (step S6-1). Then, the process proceeds to step S6-2. In the present embodiment, as the value of the combustion state determination value ion, 6 is normal combustion, a value larger than 6 is lean-side combustion, a value smaller than 6 is rich-side combustion, and 0 is misfired or unknown. .

  The AD conversion value ionad is obtained by AD-converting the combustion state determination result output from the combustion state detection device 30 in step S3 and storing the result in the variable ionad. By preparing a process for replacing the AD conversion value ionad of the state determination result with a desired value for control, replacement with a system that detects another combustion state is facilitated without changing the control content.

In the next step S6-2, the combustion state determination value ion is checked, and if it is not 0 indicating misfire or unknown, the annealing calculation shown in the following equation (1) is performed to calculate the smoothed combustion state determination value ionsm. (Step S6-3), the process is exited. Further, when the condition that ion ≠ 0 is not satisfied in step S6-2, the process is exited.
ionsm (n) = ionsm (n−1) × (1−kIOMSM) + ion × kIONSM (1)
However, kIONSM: smoothing coefficient (for example, kIONSM = 0.5)

  By performing the smoothing calculation of the combustion state determination value ion, the influence of variations in the combustion state between cylinders and the influence of partial misalignment of the fuel injection amount are alleviated, and the erroneous determination frequency of fuel properties is reduced. The reliability of the fuel property determination can be improved.

[Calculation process of combustion state determination count]
Next, “combustion state determination count calculation processing” in step S7, which is executed subsequent to “calculation of combustion state determination value” in step S6, will be described.

  In the present embodiment, for each processing unit of the combustion state detection cycle wetcycle, the threshold value indicating the lean combustion state when using heavy fuel is set in two stages (Low level, Hi level) with respect to the combustion state determination value ion. ) Threshold values tIONL_L and tIONH_L, and threshold values representing the rich combustion state when light fuel is used are set as two-level (Low level and Hi level) threshold values tIONL_R and tIONH_R. Then, for each processing unit of the combustion state detection cycle wetcycle, the number of experiences that are greater than or equal to the threshold value representing the lean combustion state or less than or equal to the threshold value representing the rich combustion state is classified and counted by each set threshold value.

Examples of setting threshold values for the combustion state determination value ion are shown in the following (a) and (b).
(A) Threshold value indicating lean combustion state when heavy fuel is used When wecycleout = 0, tIONL_L = 100
tIONH_L = 100
When wecycleout ≠ 0, tIONL_L = 8
tIONH_L = 10
(B) Threshold value indicating rich combustion state when light fuel is used When wetcycle = 0, tIONL_R = 100
tIONH_R = 100
When wecycleout ≠ 0, tIONL_R = 4
tIONH_R = 2

  Similarly, with respect to another smoothed combustion state determination value ionsm, for each processing unit of the combustion state detection cycle wetcycle, a threshold value indicating a lean combustion state when heavy fuel is used and rich combustion when light fuel is used The threshold value representing the state is set with one-stage threshold values tIONSM_L and tIONSM_R, respectively. Then, for each processing unit of the combustion state detection cycle wetcycle, the number of experiences that are greater than or equal to the threshold value representing the lean combustion state or less than or equal to the threshold value representing the rich combustion state is classified and counted.

Examples of setting the threshold value for the annealing state determination value ionsm are shown in the following (c) and (d).
(C) Threshold value indicating lean combustion state when heavy fuel is used When wecycleout = 0, tIONSM_L = 100
When wecycleout ≠ 0, tIONSM_L = 9
(D) Threshold value representing rich combustion state when light fuel is used When wetcycle = 0, tIONSM_R = 100
When wecycleout ≠ 0, tIONSM_R = 3

  The above threshold values tIONL_L, tIONH_L, tIONL_R, tIONH_R, tIONSM_L, and tIONSM_R are set from a two-dimensional map of the combustion state detection cycle wetcycle and the engine cooling water temperature when the fuel property determination pattern differs significantly depending on the environmental temperature. You may do it.

  The above is the outline of the “combustion state determination value count calculation process”, and the detailed procedure is shown in FIG. This process is executed every TDC-40 deg calculation cycle. First, the previous combustion state detection cycle wetcycle (n-1) is compared with the current combustion state detection cycle wetcycle (n), and the condition that they are not equal is established. It is determined whether or not to perform (step S7-1).

  As a result, when the condition of wetcycle (n-1) ≠ wecycle (n) is satisfied, the combustion state determination counts ionh_lcnt, ionl_lcnt, ionsm_lcnt, and the rich side are the number of times of experience on the lean side with respect to the combustion state determination values ion and ionsm. After resetting the combustion state determination counts ionh_rcnt, ionl_rcnt, and ions_rcnt, which are the number of experiences (step S7-2), the process proceeds to step S7-3. Further, when the condition in step S7-1 is not satisfied, the process proceeds from step S7-1 to step S7-3.

  In step S7-3, the combustion state determination value ion is checked, and if it is not 0 indicating misfire or unknown, the process proceeds to step S7-4. When the condition is not satisfied in step S7-3 (ion = 0), the process exits.

  In step S7-4, the combustion state determination value ion is compared with the second stage threshold value (Hi level) tIONH_L representing the lean combustion state, and if it is equal to or greater than the threshold value, the second stage (Hi level) representing the lean combustion state. The count ionh_lcnt for the threshold tIONH_L and the count ionl_lcnt for the first stage (Low level) threshold tIONL_L are respectively incremented (step S7-5), and the process proceeds to step S7-12.

  When the condition in step S7-4 is not satisfied, the combustion state determination value ion is compared with the first stage (Low level) threshold tIONL_L representing the lean combustion state (step S7-6). The count ionl_lcnt is incremented (step S7-7), and the process proceeds to step S7-12.

  When the condition in step S7-6 is not satisfied, the combustion state determination value ion is compared with the second stage (Hi level) threshold value tIONH_R representing the rich combustion state (step S7-8). The count ionh_rcnt for the second stage (Hi level) threshold tIONH_R representing the state and the count ionl_rcnt for the first stage (Low level) threshold tIONL_R are respectively incremented (step S7-9), and the process proceeds to step S7-12. .

  When the condition of step S7-8 is not satisfied, the combustion state determination value ion is compared with the first stage (Low level) threshold value tIONL_R representing the rich combustion state (step S7-10). The count ionl_rcnt is incremented (step S7-11), and the process proceeds to step S7-12. When the condition of step S7-10 is not satisfied, the process proceeds to step S7-12.

  In step S7-12, the smoothed combustion state determination value ionsm is compared with a threshold value tIONSM_L indicating the lean combustion state. If the threshold value is greater than or equal to the threshold value, the count ionsm_lcnt for the threshold value is incremented (step S7-13). Exit.

  When the condition of step S7-12 is not satisfied, the smoothed combustion state determination value ionsm is compared with the threshold value tIONSM_R representing the rich combustion state (step S7-14), and if it is equal to or less than the threshold value, the count ionsm_rcnt for the threshold value is incremented. (Step S7-15), the process is exited and the process proceeds to Step S8 in FIG. When the condition in step S7-14 is not satisfied, the process exits from the present process and proceeds to step S8 in FIG.

[Fuel property judgment process]
Next, following the “combustion state determination count calculation process” in step S7 described above, the process proceeds to step S8, where the “fuel property determination process” is executed. In this fuel property determination process, for each processing unit of the combustion state detection cycle wetcycle, the count results (combustion state determination counts) classified according to the above-described threshold values, and when heavy fuel or light fuel is used for each threshold value If the count result is equal to or greater than the count judgment value when using heavy fuel, it is determined that the fuel used is heavy fuel, and the count result is greater than or equal to the count judgment value when using light fuel. In this case, it is determined that the fuel used is light fuel.
Examples of setting each count determination value are shown in the following (1) to (6).

(1) Heavy fuel judgment value for lean side Hi level experience count (count judgment value tIONH_LCNT for count ionh_lcnt)
When wecycleout = 0, tIONH_LCNT = 100
When wecycleout ≦ 40, tIONH_LCNT = 2
When wecycle> 40, tIONH_LCNT = 3

(2) Heavy fuel judgment value for lean side Low level experience count (count judgment value tIONL_LCNT for count ionl_lcnt)
When wecycleout = 0, tIONL_LCNT = 100
When wecycleout ≦ 40, tIONL_LCNT = 3
When wecycle> 40, tIONL_LCNT = 5

(3) Heavy fuel determination value for the number of lean-side experiences after annealing (count determination value tIONSM_LCNT for count ions_lcnt)
When wecycleout = 0, tIONSM_LCNT = 100
When wetcycle ≤ 40, tIONSM_LCNT = 2
When wecycle> 40, tIONSM_LCNT = 3

(4) Light fuel judgment value with respect to the rich side Hi level experience count (count judgment value tIONH_RCNT with respect to the count ionh_rcnt)
When wecycleout = 0, tIONH_RCNT = 1
When wetcycle ≤ 40, tIONH_RCNT = 2
When wecycle> 40, tIONH_RCNT = 3

(5) Light fuel determination value with respect to the number of times of rich-side low level experience (count determination value tIONL_RCNT with respect to the count ionl_rcnt)
When wecycleout = 0, tIONL_RCNT = 1
When wetcycle ≤ 40, tIONL_RCNT = 3
When wecycle> 40, tIONL_RCNT = 5

(6) Light fuel determination value for the number of rich-side experiences after annealing (count determination value tIONSM_RCNT for countionsm_rcnt)
When wecycleout = 0, tIONSM_RCNT = 1
When wetcycle ≤ 40, tIONSM_RCNT = 2
When wecycle> 40, tIONSM_RCNT = 3

  Each of the above determination values tIONH_LCNT, tIONL_LCNT, tIONSM_LCNT, tIONH_RCNT, tIONL_RCNT, and tIONSM_RCNT is obtained from a two-dimensional map of the combustion state detection cycle wetcycle and the engine cooling water temperature when the fuel property determination pattern differs significantly depending on the environmental temperature. You may make it set.

  Specifically, the fuel property determination process is executed according to the procedure shown in FIG. 10 every TDC-40 deg calculation cycle.

  That is, first, the combustion state determination value ion is checked (step S8-1). If it is not 0 indicating misfire or unknown, the process proceeds to step S8-2. When the condition in step S8-1 is not satisfied, the normal combustion determination flag xcomb indicating normal combustion is reset (step S8-9), and then the present determination process is exited.

  In step S8-2, the counting results ionh_lcnt, ionl_lcnt, and ions_lcnt classified by the respective thresholds representing the lean combustion state when heavy fuel is used with respect to the combustion state determination values ion and ionsm are counted as the heavy fuel counts at the respective thresholds. The determination values tIONH_LCNT, tIONL_LCNT, and tIONSM_LCNT are respectively compared. If it is equal to or greater than the present count determination value, it is determined that the fuel used is heavy fuel, 3 is set in the fuel property determination value fuel (step S8-3), and the process proceeds to step S8-7.

  When the condition in step S8-2 is not satisfied, the counting results ionh_rcnt, ionl_rcnt, and ions_rcnt classified by the respective threshold values representing the rich combustion state when the light fuel is used with respect to the combustion state determination values “ion” and “ionssm” are determined as the light fuel counts at the respective threshold values. The values tIONH_RCNT, tIONL_RCNT, and tIONSM_RCNT are respectively compared (step S8-4). If it is equal to or greater than this count determination value, it is determined that the fuel used is light fuel, 1 is set to the fuel property determination value fuel (step S8-5), and the process proceeds to step S8-7. When the condition in step S8-4 is not satisfied, the previous fuel property is maintained, the normal combustion determination flag xcomb is set (step S8-6), and the process is exited.

  In step S8-7, when the combustion state detection cycle wetcycle is other than 0, the previous fuel property determination value fuel (n-1) is compared with the current fuel property determination value fuel (n), and the heavy fuel is detected. After changing to light fuel or from light fuel to heavy fuel, after setting the fuel property judgment value fuel to 2 to make the fuel property an intermediate fuel (step S8-8), after resetting the normal combustion judgment flag xcomb (Step S8-9), the process is exited. When the condition in step S8-7 is not satisfied, the normal combustion determination flag xcomb is reset (step S8-9), and the process is exited.

  The condition of the fuel property determination value fuel in the above step S8-7 is a condition for recognizing whether or not the previous fuel property determination result is an erroneous determination, and performing a return when it is an erroneous determination. That is, when it is determined that the fuel property is heavy fuel, the correction works in the direction of increasing the fuel injection amount. As a result of the increase correction, if it is determined that the fuel is light, the correction is not appropriate, so the fuel property is determined as intermediate fuel again. On the contrary, similarly, when it is determined that the fuel property is light fuel, the correction works in the direction of decreasing the fuel injection amount. As a result of the reduction correction, if it is determined that the fuel is heavy, the correction is not appropriate, so the fuel property is determined as intermediate fuel again.

  Also, the condition of “wetcycleout ≠ 0” in step S8-7 is that when the combustible air is filled upstream of the intake system due to the oil-tight leak of the injector and combustion is caused by invalid ignition, the light fuel is selected to avoid over-richness. It is a condition to do. That is, wetcycle = 0 indicates the “ignition ignition combustion state determination result storage period”, and a combustion state determination result other than 0 representing misfire or unknown is obtained from the combustion state detection device 30 during the period of wetcycle = 0. In other words, when combustible air is filled upstream of the intake system due to oil-tight leakage of the injector and combustion is caused by reactive ignition, regardless of the learned value of fuel properties, the light fuel with the smallest amount of fuel injection is selected. Avoid rich.

  As described above, in the processes of steps S7 and S8, a plurality of threshold values for classifying the combustion state determination values and the count determination values for these threshold values are appropriately set for each processing unit of the combustion state detection cycle wetcycle. Thus, a fuel property determination pattern suitable for each processing unit of the combustion state detection cycle wetcycle can be created, and the influence of variations in the combustion state between the cylinders can be mitigated, and in one processing unit of the combustion state detection cycle wetcycle. Accurate fuel properties can be determined.

  In addition, even if an erroneous determination of the fuel property occurs due to a partial misalignment of the fuel injection amount in the early stage of engine starting where a significant difference in the fuel injection amount occurs depending on the fuel property, for each processing unit of the combustion state detection cycle wetcycle. Since the fuel property determination is performed a plurality of times, it is possible to recover from the erroneous determination and improve the reliability of the fuel property determination result.

[Calculation processing of injection amount magnification]
Next, the calculation process of the injection amount magnification tauamp in step S9 will be described. In the present embodiment, heavy fuel and light fuel injections are performed for each processing unit of the port wet cycle wetcycle, with the fuel injection amount when the intermediate fuel is used as a reference (intermediate fuel injection amount magnification tauamp is 1.0). The amount magnification is set in advance, and the injection amount magnification is calculated based on the fuel property determined in the previous step S8, so that a remarkable difference in the fuel injection amount occurs depending on the fuel property. Enables correct fuel injection amount correction.

  Further, when the fuel property determined in step S8 is heavy fuel, the fuel amount is gradually increased between the heavy fuel injection amount magnification and the intermediate fuel injection amount magnification by using an injection amount magnification attenuation coefficient tauampdmp described later. An annealing calculation is performed to increase the reflection rate of the intermediate fuel injection magnification. At the time of convergence to the injection amount magnification of the intermediate fuel, the fuel property is updated to the intermediate fuel.

  The above is the outline description of “calculation of the injection amount magnification”, and the detailed procedure is performed according to the flowchart shown in FIG. This process is performed every TDC-40 deg calculation cycle, and first, the combustion state determination value ion is checked (step S9-1). And if it is not 0 showing misfire or unknown, it will transfer to Step S9-2, and if the conditions of Step S9-1 are not satisfied, it will transfer to Step S9-8.

  In step S9-2, the condition that the “normal combustion determination flag is in the set state (xcomb = 1)” and the “fuel property is heavy fuel (fuel = 3)” is checked. When the condition of step S9-2 is not satisfied, the injection amount magnification attenuation coefficient taupampdmp is set to 1.0 (step S9-7), and the process proceeds to step S9-8.

  On the other hand, when the condition of step S9-2 is satisfied, the injection amount magnification attenuation coefficient taupampdmp is subtracted by the set value tTAUAMPDMP (step S9-3), and the subtraction result is checked (step S9-4). The set value tTAUAMPDMP is set for each processing unit of the port wet cycle. For example, under the condition of wetcycle ≦ 10, it is set to tTAUAMPPDMP = 0.125, and under the condition of wetcycle> 10, tTAUAMPPDMP is set to 0.084. Is set.

  When the heavy fuel injection amount magnification and the light fuel injection amount magnification change significantly depending on the environmental temperature, the set value tTAUAMPPDMP is set from a two-dimensional map of the port wet cycle wetcycle and the engine coolant temperature. Anyway.

  If the subtraction result of the injection amount magnification attenuation coefficient tauampdmp is 0 or less in the check in step S9-4, the injection amount magnification attenuation coefficient tauampdmp is set to 1.0 (step S9-5), and the fuel property determination value fuel is set. The fuel is updated to intermediate fuel (fuel = 2) (step S9-6), and the process proceeds to step S9-8. When the condition of step S9-4 is not satisfied, the process proceeds to step S9-8.

In step S9-8, it is checked whether the fuel property is heavy fuel (fuel = 3). As a result, in the case of heavy fuel, injection is performed by performing a smoothing calculation using the injection amount magnification attenuation coefficient tauampdmp as shown in the following equation (2) for the heavy fuel injection amount magnification tHEAVYAMP set in advance. A quantity magnification taupamp is calculated (step S9-9), and the process is exited.
tauamp = 1.0 × (1−tauampdmp) + tHEAVYAMP × tauampdmp (2)

  The injection amount magnification tHEAVYAMP preset for the heavy fuel is preset for each port wet cycle wetcycle, with the injection amount magnification of the intermediate fuel being 1.0, as shown by the solid line in FIG. When the heavy fuel injection amount magnification changes significantly depending on the environmental temperature, the injection amount magnification tHEAVYAMP may be set from a two-dimensional map of the port wet cycle wetcycle and the engine coolant temperature.

  On the other hand, when the condition of step S9-8 is not satisfied, it is checked whether the fuel property is intermediate fuel (fuel = 2) (step S9-10). If it is intermediate fuel, the injection amount magnification attenuation coefficient taupampdmp is set to 1.0. Is set (step S9-11), and the process exits. When the condition of step S9-10 is not satisfied, the light fuel injection amount magnification is set to a preset injection amount magnification tLIGHTAMP (step S9-12), and the process is exited.

  As shown by the broken line in FIG. 14, the injection amount magnification tLIGHTHTAMP set in advance for the light fuel is set in advance for each port wet cycle wetcycle with the intermediate fuel injection amount magnification being 1.0. If the light fuel injection amount magnification changes significantly depending on the environmental temperature, the injection amount magnification tLIGHTAMP may be set from a two-dimensional map of the port wet cycle wetcycle and the engine coolant temperature.

  In the above injection amount magnification calculation processing, in the case of heavy fuel with the largest fuel injection amount, control is performed in a direction to always decrease the injection amount using the injection amount magnification attenuation coefficient taupampdmp. Even if the fuel properties of the fuel used do not completely match, the fuel injection amount can be corrected to the optimum amount. Further, by updating the fuel property to the intermediate fuel when the injection amount magnification attenuation coefficient tauampdmp reaches 0.0, that is, when it converges to the fuel injection amount of the intermediate fuel, the fuel property from the heavy fuel to the intermediate fuel. The judgment pattern is used.

  Further, the subtraction amount of the injection amount magnification attenuation coefficient tauampdmp in step S9-3 can be set for each processing unit of the port wet cycle wetcycle, and the subtraction amount commensurate with the fuel injection amount can be arbitrarily set. it can. As shown in FIG. 16, within the fuel property learning value update period (the fuel property learning value update period will be described later), the subtraction amount is set to be relatively large so that the injection amount magnification attenuation coefficient taupampmp becomes 0.0. .

  In other words, the speed at which the reflection rate of the intermediate fuel injection amount magnification is increased can be changed for each processing unit of the port wet cycle wetcycle, and the fuel injection amount is reduced by setting it relatively early at the start of the engine. It is possible to improve the determination accuracy by shortening the time required to perform the determination and performing the fuel property determination at the initial stage of engine start in which a significant difference in the fuel injection amount occurs depending on the fuel property.

  FIGS. 16 to 18 show an example in which the fuel property learning value fuel_buram read when the ignition switch is turned on is heavy fuel, so that the subtraction process of the injection amount magnification attenuation coefficient taupampmp is executed.

  In the example shown in FIG. 16, as a result of the subtraction until the injection amount magnification attenuation coefficient taupampmp reaches 0.0, the fuel property is updated to intermediate fuel, and 1.0, which is the injection amount magnification attenuation coefficient of the intermediate fuel, is set. The In the example of FIG. 17, it is determined that the fuel property is heavy fuel again in the subtraction process of the injection amount magnification attenuation coefficient taupampmp. In this case, the injection amount magnification attenuation coefficient tauampdmp behaves like a PI correction, and the fuel injection amount also exhibits the same behavior. In the example of FIG. 18, since there is no significant difference in the fuel injection amount depending on the fuel property outside the fuel property learning value update period, the subtraction amount of the injection amount magnification attenuation coefficient tauampdmp is set small. The

[Learning value writing (backup RAM writing) processing]
In the main process of FIG. 2, following the above step S9, a process of writing the learning value fuel_buram to the backup RAM in step S10 is executed.

  Details of this “backup RAM writing process” are shown in FIG. 12, and “the engine speed is a set value kNEBU (for example, kNEBU = 300) or more” for each TDC-40 deg calculation cycle, and “combustion state detection cycle” A condition that “wetcycleout is a set value kWETBU (for example, kWETBU = 10) or less” and “no experience of state change” is checked (step S10-1). When the condition is satisfied in step S10-1, the process proceeds to step S10-2. When the condition is not satisfied, the process is exited.

  In step S10-2, a condition that “the injection amount magnification attenuation coefficient tauampdmp is greater than a set threshold value kFUEL2 (eg, kFUEL2 = 0.2)” is checked. When the condition is satisfied, the current fuel property fuel is written in the backup RAM as a learned value fuel_buram (step S10-3), and the process is exited. When the condition is not satisfied, the learned value fuel_buram (fuel_buram = 2) of the intermediate fuel is written in the backup RAM (step S10-4), and the process is exited.

  In order to accurately determine the fuel properties, the fuel properties must be limited to the initial period when the fuel injection amount varies significantly depending on the fuel properties. Therefore, in step S10-1, the learning value update period of the fuel property is limited by the condition of the combustion state detection cycle wetcycle. The rotational speed condition is for excluding the engine cranking period, which is the initial operation of engine start.

  The “state change” of the “no state change experience” condition in step S10-1 means that when these load states change from an unloaded state such as an electric load, an air conditioner, and a power steering before the engine starts, before the engine starts. When the idle state is released from the idle state by the accelerator operation, the shift position of the transmission before starting the engine is operated from the neutral state to the drive position.When the state change occurs, the fuel property In order to reduce the fuel property determination accuracy other than the fuel injection amount correction based on it, the update of the learning value is prohibited when a state change is experienced.

  Further, the condition that “the injection amount magnification attenuation coefficient taupampdmp is equal to or less than the set threshold value kFUEL2” in step S10-2 is a condition for storing and holding as a learning value when the fuel property is close to the preset intermediate fuel. FIG. 19 re-presents an example in which the fuel property learning value fuel_buram read when the ignition switch is turned on is heavy fuel, and thus the subtraction processing of the injection amount magnification attenuation coefficient taupampmp is executed.

  In this example, in the subtraction process of the injection amount magnification attenuation coefficient tauampdmp, once it is determined that the fuel property is heavy fuel, it is subtracted again until it is close to 0.0. In the case where the injection amount magnification attenuation coefficient taupampdmp is close to 0.0 within the fuel property learning value update period, the injection amount magnification attenuation coefficient tauampdmp is set in advance in accordance with the condition “the injection amount magnification attenuation coefficient tauampdmp is not more than the set threshold value kFUEL2” in step S10-2. It is determined that the fuel property is close to the intermediate fuel, and is stored as a learning value.

[Injection amount calculation processing]
Next, “calculation of injection amount” in step S11 of the main process in FIG. 2 will be described. The details of this injection amount calculation process are shown in FIG. 13, and the condition “Fuel property determination execution condition is satisfied” is checked every TDC calculation cycle (step S11-1). When the condition of step S11-1 is satisfied, the fuel injection amount tauout is calculated by the calculation of the following equation (3) (step S11-2), and the process is exited.
tauout = tau × tauamp (3)
However, tauout: fuel injection amount after fuel property correction
tau: fuel injection amount of intermediate fuel
tauamp: Injection rate magnification

The fuel injection amount tauout may be expressed by the following equation (4) instead of equation (3).
tauout = tp × (Ffb + Fln) × {1+ (Fas + Ftw + Fda) × tauamp +...} + tv (4)
Where tauout: corrected fuel injection amount
tp: Basic injection amount
Ffb: Air-fuel ratio feedback correction coefficient
Fln: Air-fuel ratio correction learning value
Fas: Increase correction coefficient after start of intermediate fuel
Ftw: Water temperature increase correction factor for intermediate fuel
Fda: Intermediate fuel acceleration increase correction coefficient
tauamp: Injection rate magnification
tv: Voltage correction

  When the fuel property determination execution condition in step S11-1 is not satisfied, that is, when the air-fuel ratio feedback control is started, the fuel injection amount correction by the air-fuel ratio feedback control overlaps with the fuel injection amount correction based on the fuel property. To avoid complication, standardize fuel injection amount correction by air-fuel ratio feedback control. In order not to affect the controllability of the air-fuel ratio feedback control, the fuel injection amount correction based on the fuel property, that is, the injection amount magnification is gradually changed to converge to a fixed value (1.0).

  Specifically, the injection amount magnification tauamp is checked (step S11-3), and if it is 1.0 or more, the set value kTAUDMP (for example, kTAUDMP = 0.02) is subtracted from the injection amount magnification tauamp (step k11). S11-4). After subtraction in step S11-4, the injection amount magnification tauamp is checked again (step S11-5). If smaller than 1.0, 1.0 is set in the injection amount magnification tauamp (step S11-6). ) And (3) are calculated to calculate the fuel injection amount tauout (step S11-2), and the process is exited.

  When the condition in step S11-3 is not satisfied, the injection amount magnification tauamp is added by the set value kTAUDMP (step S11-7), and then the injection amount magnification tauamp is checked again (step S11-8). When the addition result is 1.0 or more, after setting 1.0 to the injection amount magnification tauamp (step S11-6), the fuel injection amount tauout is calculated by the calculation of the equation (3) (step S11-). 2) Exit this process. When the condition of step S11-5 or step S11-8 is not satisfied, the fuel injection amount tauout is calculated by the calculation of equation (3) (step S11-2), and the process is exited.

Configuration diagram of engine control system Flow chart showing main processing Initial processing flowchart Flow chart of ignition number calculation process Flow chart of combustion state determination result acquisition processing Flow chart of fuel property determination execution condition establishment determination process Port wet cycle calculation flowchart Flow chart of combustion state determination value calculation processing Flow chart of combustion state determination count calculation processing Flow chart of fuel property determination process Flow chart of injection amount magnification calculation processing Flow chart of learning value writing process Flow chart of injection amount calculation process Explanatory diagram showing the injection amount magnification of heavy fuel and light fuel based on the fuel injection amount of intermediate fuel Explanatory drawing which shows the relationship between the output timing of a combustion state determination result, and each processing timing Explanatory drawing which shows the behavior of the injection quantity magnification attenuation coefficient when heavy fuel is not judged at the time of the fuel property judgment pattern from heavy fuel to intermediate fuel Explanatory drawing which shows the behavior of the injection quantity magnification attenuation coefficient in the case of repeating heavy fuel determination at the time of fuel property determination pattern from heavy fuel to intermediate fuel Explanatory drawing which shows the behavior of the injection amount magnification attenuation coefficient when the subtraction amount is reduced outside the fuel property learning value update period in the fuel property determination pattern from heavy fuel to intermediate fuel Explanatory drawing which shows the behavior of the injection quantity magnification attenuation coefficient when it is subtracted to near zero within the fuel property learning value update period in the fuel property determination pattern from heavy fuel to intermediate fuel Explanatory drawing showing the relationship between port wet cycle and combustion state detection cycle

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Engine 30 Combustion state detection apparatus 50 Engine control apparatus 51 Port wet detection part 52 Fuel property determination part 53 Fuel injection amount correction | amendment fuel Fuel property determination value fuel_buram Fuel property learning value ion, ionsm Combustion state determination value taampamp Injection amount magnification taauampdmp injection Amount magnification attenuation coefficient tauout Fuel injection amount wetcycle port wet cycle wetcycle combustion state detection cycle

Claims (20)

A port wet detection unit for detecting a port wet that changes according to the fuel injection amount and the number of injections of the fuel injected into the intake system;
A fuel property determination unit that determines the fuel property of the fuel used from the combustion state in association with the port wet detected by the port wet detection unit;
A fuel injection amount correction unit that corrects the fuel injection amount in association with the port wet detected by the port wet detection unit from the fuel property determined by the fuel property determination unit ;
The port wet detection unit
Defining a port wet cycle in which the number of ignitions or number of injections with the same port wet is one processing unit, and detecting the port wet for each processing unit of the port wet cycle;
The fuel property determination unit
A combustion state detection cycle in which a period for detecting the combustion state of one processing unit of the port wet cycle is further defined as one processing unit, and each processing unit of the port wet cycle and each processing unit of the combustion state detection cycle is defined. And determining the fuel property for each processing unit of the combustion state detection cycle . 2. The period before the start of the combustion state detection cycle is a period for detecting a combustion state of ignition for a cylinder in which fuel injection is not performed during engine cranking, which is an initial operation of engine start. Engine control device. The fuel property determination unit
The number of times that the combustion state judgment value that expresses the combustion state numerically experiences a value that is greater than or equal to the threshold value that represents the lean combustion state when using heavy fuel, and the number of times that experience less than or equal to the threshold value that represents the rich combustion state when using light fuel Are counted for each processing unit of the combustion state detection cycle,
For each processing unit of the combustion state detection cycle, if the number of experiences equal to or greater than the threshold value representing the lean combustion state when using heavy fuel is greater than or equal to the heavy fuel count determination value, it is determined that the fuel is heavy fuel, 3. The engine control according to claim 1, wherein if the number of experiences less than or equal to a threshold value representing a rich combustion state when using light fuel is equal to or greater than a light fuel count determination value, the engine is determined to be light fuel. apparatus. After the engine is started, the result of the smoothing calculation performed between the initial value before the engine start and the combustion state determination value is used as the new combustion state determination value, and the previous smoothing calculation result and the current combustion state are sequentially calculated. The engine control apparatus according to claim 3 , wherein a result obtained by performing the smoothing calculation with the state determination value is set as a new combustion state determination value. The engine control device according to claim 3 or 4 , wherein a plurality of threshold values representing a lean combustion state when using heavy fuel and a plurality of threshold values representing a rich combustion state when using light fuel are provided. The fuel injection amount correction unit is
Based on the fuel injection amount when using intermediate fuel, which shows the average properties of heavy fuel and light fuel, the injection amount magnification of heavy fuel and light fuel is set for each processing unit of the port wet cycle. The engine control apparatus according to any one of claims 3 to 5 , wherein the fuel injection amount is corrected based on an injection amount magnification according to the fuel property determined by the fuel property determination unit. The engine control according to claim 6 , wherein when it is determined that the fuel property has changed from heavy fuel to light fuel or from light fuel to heavy fuel, the fuel property is re-determined as intermediate fuel. apparatus. The fuel property determination unit determines that the fuel property is light fuel when a combustion state determination value other than misfire is obtained regardless of the value of the combustion state determination value before the start of the combustion state detection cycle. The engine control device according to claim 6 . When the fuel property determined by the fuel property determination unit is heavy fuel, gradually increase the reflection ratio of the intermediate fuel injection amount magnification between the heavy fuel injection amount magnification and the intermediate fuel injection amount magnification. The engine control apparatus according to any one of claims 6 to 8 , wherein a smoothing calculation is performed, and the fuel injection amount is corrected by an injection amount magnification obtained as a result of the smoothing calculation. The engine control apparatus according to claim 9 , wherein the fuel property is updated to the intermediate fuel when it converges to the injection amount magnification of the intermediate fuel. The engine coolant temperature immediately before starting the engine, said port wet detection unit, the fuel property determination unit, and any one of claims 1-10, characterized in that to set as the execution condition of the fuel injection amount correction section The engine control apparatus described in 1. The engine control device according to claim 11 , wherein the port wet detection unit, the fuel property determination unit, and the fuel injection amount correction unit are implemented before the start of air-fuel ratio feedback control. . According to claim 12, characterized in that for converging the injection amount ratio obtained to the air-fuel ratio feedback control immediately before the start, to a fixed value by gradually changing the air-fuel ratio feedback control starting later by the fuel injection amount correction unit Engine control device. The engine control device according to any one of claims 1 to 13 , wherein the learning value of the fuel property is read out and used immediately before starting the engine. The range of the engine coolant temperature immediately before starting the engine is set as a condition that reflects the learned value of the fuel property, and when the temperature is outside the set temperature range, the fuel injection amount correction by the fuel injection amount correction unit is not performed. The engine control device according to claim 14 . If there is no experience in rewriting the learned value of the fuel property, heavy fuel is set for the fuel property, and if the learned value of the fuel property is light fuel, the average property of the heavy fuel and the light fuel is used for the fuel property. The engine control apparatus according to claim 14 , wherein an intermediate fuel is set. The engine control device according to any one of claims 14 to 16 , wherein a period during which the fuel property learning value is updated is limited to an initial period of engine start. It is possible to set the rate of increasing the reflection rate of the injection amount magnification of the intermediate fuel, which shows the average characteristics of heavy fuel and light fuel, for each processing unit of the port wet cycle, The engine control device according to claim 17 , wherein the engine control device is set quickly. 19. The engine control device according to claim 17 , wherein when a change in an operating state is experienced after the engine is started, updating of the learning value of the fuel property is prohibited. If the injection amount magnification converges within the set range for the intermediate fuel indicating the average characteristics of heavy fuel and light fuel within the update period of the fuel property learning value, the intermediate fuel is The engine control device according to any one of claims 17 to 19 , wherein the engine control device is stored as a learning value.

Images