JP2018193948A - Control method and device for engine, and method and device for detecting the number of discharge particles in engine - Google Patents

Abstract

To suppress the number of particulate substances (PN) discharged from an engine.SOLUTION: A control method for engine comprises: causing an optical sensor SN3 to detect luminous flame that is light emission from particulate substances existing in a cylinder 2; analyzing the detected luminous flame with the two-color method to obtain a KL value; and injecting fuel from an injector 15 in such a mode that in a case where the obtained KL value is large, the number of particulate substances is suppressed compared to a case where the KL value is small. The control method comprises, for example, controlling fuel injection from the injector 15 so that in the case where the KL value is large, an adhesion amount of fuel to a wall surface of a piston 5 is reduced compared to the case where the KL value is small.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for suppressing the number of particulate matter discharged from an engine.

  For example, an engine that injects fuel into a cylinder, such as a gasoline direct injection engine, is known to generate more particulate matter such as soot than conventional gasoline engines that employ port injection. Yes. In particular, under conditions where the catalyst temperature is low, such as immediately after the engine is cold started, control (catalyst warm-up operation) is performed to raise the temperature of the exhaust gas by, for example, retarding the ignition timing. It is known that many particulate substances are generated during the catalyst warm-up operation.

  In order to suppress the particulate matter, it is necessary to accurately grasp the generation amount, and in Patent Document 1 below, a combustion analysis technique for that purpose is proposed. Specifically, in Patent Document 1, radiation light from a combustion flame in a cylinder is detected, a KL value is calculated by a so-called two-color method, and a soot generation amount is calculated by a predetermined calculation using the KL value. Seeking.

JP-A-2015-99055

  On the other hand, as a recent trend, there is a movement to regulate not only the mass of particulate matter discharged from the engine but also the number of particulate matter (hereinafter referred to as PN as appropriate), and technology development to meet this PN regulation Is required.

  However, with the technique of the above-mentioned Patent Document 1, although the amount of soot generation (mass) in the cylinder can be accurately predicted to some extent, the number of particulate matter containing soot discharged from the engine, that is, the PN is accurately determined. It is difficult to predict. That is, since particulate matter includes various sizes of small and large particles, it is difficult to accurately predict the PN discharged from the engine only by knowing the soot mass generated in the cylinder.

  The present invention has been made in view of the circumstances as described above, and a technique capable of accurately predicting PN discharged from an engine, or a technique capable of effectively suppressing PN by control based on the prediction. The purpose is to provide.

  As a result of intensive research conducted by the inventors of the present invention to solve the above-mentioned problems, the KL value obtained by analyzing the luminous flame, which is the emission of particulate matter existing in the cylinder, by the two-color method is the particle discharged from the engine. It was found that there is a high correlation with PN, which is the number of particulate substances, and that the larger the KL value, the more PN. And based on such knowledge, the following invention was completed.

  First, a first invention of the present application is a method for controlling an engine having an injector for injecting fuel into a cylinder, and optically detects a luminous flame which is emission of particulate matter existing in the cylinder. The first step, the second step of analyzing the luminous flame detected in the first step by a two-color method to obtain the KL value, and the KL value obtained in the second step being large Includes a third step of injecting fuel from the injector in a mode in which the number of the particulate matter is suppressed as compared with a case where the number is small (Claim 1).

  According to the first aspect of the present invention, when the KL value obtained by analyzing the luminous flame in the cylinder by the two-color method is large, that is, it is predicted that there is a large PN that is the number of particulate matter discharged from the engine. In this case, since the mode of fuel injection is switched in a direction in which PN is suppressed, combustion can be properly controlled so that excessive particulate matter is not generated, and PN discharged from the engine is effectively suppressed. can do.

  In the first aspect of the invention, preferably, in the third step, the amount of fuel adhering to the piston wall surface is reduced when the KL value obtained in the second step is large compared to when the KL value is small. Thus, the fuel injection from the injector is controlled (Claim 2).

  As described above, when the amount of fuel adhering to the piston wall surface is reduced when the KL value is large (that is, when it is predicted that the PN is large), the droplet fuel adhering to the piston wall surface is reduced. Since the pool combustion which arises when a flame reaches | attains can be suppressed, generation | occurrence | production of the particulate matter resulting from the said pool combustion can be suppressed, and PN can be suppressed effectively.

  In the above-described configuration, more preferably, the injector is controlled to inject fuel at a plurality of timings including a compression stroke during a catalyst warm-up operation in which the temperature of the catalyst is low. In the third step, When the KL value obtained in step 2 is large, the fuel injection pressure from the injector is increased as compared to when the KL value is small (Claim 3).

  During the catalyst warm-up operation, in general, in order to raise the temperature of the exhaust gas, for example, measures such as retarding the ignition timing of the air-fuel mixture are taken, so it can be said that the combustion stability is likely to be lowered. On the other hand, in the said structure, since at least one part of fuel is injected during a compression stroke, stratification of a fuel can be implement | achieved and the situation where combustion stability tends to fall can be improved. However, the fact that the fuel is injected during the compression stroke means that the amount of fuel that adheres as droplets to the piston wall surface increases. For this reason, depending on the case, there is a concern that the adhering fuel is so-called pool burning, and particulate matter is excessively generated. On the other hand, in the above configuration, when the KL value is large (that is, when it is predicted that the PN is large), the fuel injection pressure (fuel pressure) from the injector is increased, so that atomization of the fuel can be promoted. The time required for the fuel to vaporize and atomize can be shortened. Thereby, the fuel adhesion amount to the wall surface mentioned above can be reduced, and PN can be suppressed effectively.

  When the fuel is injected at a plurality of timings including the compression stroke during the catalyst warm-up operation as described above, in the third step, when the KL value obtained in the second step is large The amount of fuel injected at the latest timing during the compression stroke may be reduced and the amount of fuel injected before the fuel may be increased as compared with the case where the fuel is small.

  Thus, when the amount of fuel injected at the latest timing during the compression stroke is reduced, the amount of fuel that adheres as droplets to the piston wall surface can be further reduced, and PN can be effectively reduced. Can be suppressed.

  The second invention of the present application is an apparatus for controlling an engine having an injector for injecting fuel into a cylinder, and optically detects a luminous flame that is emission of particulate matter present in the cylinder. An optical sensor that performs a two-color analysis of the luminous flame detected by the optical sensor and obtains a KL value, and if the KL value obtained by the computation unit is large, compared to a case where the KL value is small, And an injection control unit that injects fuel from the injector in a mode in which the number of particulate substances is suppressed.

  According to the second invention, the same effect as that of the first invention can be obtained.

  A third invention of the present application is a method for detecting the number of particulate matter discharged from an engine, and optically detects a luminous flame which is emission of particulate matter present in the cylinder of the engine. A first step of analyzing the luminous flame detected in the first step by a two-color method to obtain a KL value, and a KL value obtained in the second step. A third step of estimating the number of particulate matter, wherein in the third step, it is estimated that the larger the KL value, the greater the number of particulate matter discharged from the engine. It is a characteristic (claim 6).

  According to this configuration, it is estimated that the larger the KL value obtained by analyzing the luminous flame in the cylinder by the two-color method, the greater the number of particulate matter discharged from the engine, that is, the PN. Based on the knowledge of the present inventors who have found a high correlation with PN, PN can be estimated easily and accurately.

  The method for obtaining the KL value by the two-color method is specifically as follows. That is, in the second step, the brightness of two different wavelength components obtained by splitting the radiation light from the luminous flame is measured, and the KL value is determined based on the brightness temperature specified from each brightness value. (Claim 7).

  Here, according to the research of the present inventor, PN particularly shows a high correlation with the peak value of the KL value obtained for each crank angle. For this reason, in the third step, it is preferable to estimate the number of the particulate matter based on the peak value of the KL value.

  According to a fourth aspect of the present invention, there is provided an apparatus for detecting the number of particulate matter discharged from an engine, and optically detecting a luminous flame that is light emission of the particulate matter present in the cylinder of the engine. An optical sensor that analyzes the luminous flame detected by the optical sensor using a two-color method to obtain a KL value, and an arithmetic unit that estimates the number of the particulate matter based on the obtained KL value, The calculation unit estimates that the larger the KL value is, the larger the number of the particulate matter discharged from the engine is (Claim 9).

  According to the fourth aspect of the invention, the same effect as that of the third aspect of the invention can be obtained.

  As described above, according to the present invention, it is possible to provide a technique capable of accurately predicting PN discharged from the engine or a technique capable of effectively suppressing PN by control based on the prediction.

1 is a system diagram schematically showing an overall configuration of an engine according to an embodiment of the present invention. It is a figure which shows roughly the spray of the fuel injected in the latter half of the compression stroke in the said engine, (a)-(c) has each shown the motion of the fuel spray injected from the different nozzle hole of an injector. . It is the first half part of the flowchart which shows the procedure of the catalyst warm-up control of the engine. It is the latter half part of the said flowchart. It is a figure for demonstrating the three injection patterns selected during the said catalyst warm-up control, (a) is the injection pattern 1, (b) is the injection pattern 2, (c) is the injection pattern 3. It corresponds. It is a figure which shows the structure of the engine for experiment used in the research used as the foundation of this invention. It is a graph which shows simultaneously the relationship between the start time (2nd * SOI) of 2nd stage fuel injection, and PN, and the relationship between the injection start time and a combustion fluctuation rate. It is a graph which shows the particle size distribution of the particle | grains made into measurement object in FIG. It is the figure which showed together the graph obtained by measuring in-cylinder pressure and a heat release rate, and the combustion flame image acquired in several crank angles. It is the figure which showed the spray scattered light image by the fuel injection of the 2nd step. FIG. 10 is a graph showing flame temperature, KL value, and luminance when two-color method measurement is performed under the same conditions as in FIG. 9, where (a) shows the average flame temperature at each crank angle, and (b) shows each crank. The KL _θ that is the KL value of the entire cylinder at the corner and the luminance are shown. Start timing of the second stage of fuel injection while (2nd · SOI) was variously changed peak value of KL _theta and (KL _.theta.max) is a graph showing the results of the PN simultaneously measured. It is a graph which shows the measurement result of KL ( _theta ) obtained in four experimental examples (comparative example and Examples 1-3) from which injection conditions differ, and the heat release rate, (a) is a waveform of KL ( _theta ), (b) Shows the waveform of the heat release rate. It is a figure which shows the distribution image of KL value obtained in the said four experiment examples. It is a graph which shows the fuel adhesion amount to the piston wall surface in the said four experiment examples. It is a graph which shows the result of having measured PN in the actual running test of vehicles, (a) shows sequential PN generated in each time, and (b) shows total PN which totaled PN of each time, respectively. ing.

(1) Description of Embodiment (1-1) Overall Configuration of Engine FIG. 1 is a system diagram schematically showing an overall configuration of an engine according to an embodiment of the present invention. The engine shown in the figure is a four-cycle gasoline direct injection engine mounted on a vehicle as a driving power source, and includes an engine body 1, an intake passage 28 through which intake air introduced into the engine body 1 circulates, And an exhaust passage 29 through which exhaust gas discharged from the engine body 1 flows.

  The engine body 1 includes a cylinder block 3 in which a cylinder 2 is formed, a cylinder head 4 attached to the upper surface of the cylinder block 3 so as to close the cylinder 2 from above, and a reciprocating slide on the cylinder 2. And an inserted piston 5. The engine main body 1 is typically of a multi-cylinder type having a plurality of (for example, four) cylinders, but here, for the sake of simplification, the description will be focused on only one cylinder 2.

  A combustion chamber 6 is defined above the piston 5, and fuel mainly composed of gasoline is supplied to the combustion chamber 6 by injection from an injector 15 described later. The supplied fuel burns while being mixed with air in the combustion chamber 6, and the piston 5 pushed down by the expansion force due to the combustion reciprocates in the vertical direction.

  Below the piston 5, a crankshaft 7 that is an output shaft of the engine body 1 is provided. The crankshaft 7 is connected to the piston 5 via a connecting rod 8 and is rotationally driven around the central axis according to the reciprocating motion (vertical motion) of the piston 5.

  The cylinder block 3 is provided with a crank angle sensor SN1 that detects the rotation angle (crank angle) of the crankshaft 7 and the rotation speed (engine rotation speed) of the crankshaft 7.

  The cylinder head 4 is provided with an intake port 9 and an exhaust port 10 that open to the combustion chamber 6, and an intake valve 11 and an exhaust valve 12 that open and close the ports 9 and 10. The intake valve 11 and the exhaust valve 12 are driven to open and close in conjunction with the rotation of the crankshaft 7 by valve mechanisms 13 and 13 including a pair of camshafts and the like disposed in the cylinder head 4.

  The intake passage 28 is connected to one side surface of the cylinder head 4 so as to communicate with the intake port 9. Air (fresh air) taken from the upstream end of the intake passage 28 is introduced into the combustion chamber 6 through the intake passage 28 and the intake port 9.

  A throttle valve 30 for adjusting the amount of intake air to the cylinder 2 is provided in the intake passage 28 so as to be openable and closable.

  The exhaust passage 29 is connected to the other side of the cylinder head 4 so as to communicate with the exhaust port 10. Burned gas (exhaust gas) generated in the combustion chamber 6 is discharged to the outside through the exhaust port 10 and the exhaust passage 29.

  A catalytic converter 31 is provided in the exhaust passage 29. The catalytic converter 31 contains a catalyst (for example, a three-way catalyst) for purifying harmful components (CO, HC, NOx, etc.) contained in the exhaust gas flowing through the exhaust passage 29. Further, the catalytic converter 31 is provided with a catalyst temperature sensor SN2 for detecting the temperature of the catalyst inside the catalytic converter 31.

  A pent roof type ridge formed on the crown surface of the piston 5 so that the central portion in the intake / exhaust direction (left-right direction in FIG. 1) is the highest, and the height decreases as the distance from the center toward the intake and exhaust sides increases. Part 5a is formed. A cavity C is formed in the center of the raised portion 5a. The cavity C is recessed on the side opposite to the cylinder head 4 (downward).

  The cylinder head 4 is provided with an injector 15 that injects fuel (gasoline) into the combustion chamber 6 and an ignition plug 16 that forcibly ignites a mixture of fuel and air injected into the combustion chamber 6.

  A fuel supply pipe 20 is connected to the injector 15, and fuel pumped from a supply pump 21 is supplied to the fuel supply pipe 20 as needed. The injector 15 injects the fuel supplied from the supply pump 21 through the fuel supply pipe 20 into the combustion chamber 6 at an appropriate timing.

  The supply pump 21 is a hydraulic or electric pump capable of changing the fuel supply pressure. Based on the set pressure in the supply pump 21, the fuel injection pressure (fuel pressure) from the injector 15 can be changed.

  The injector 15 is a six-hole type injector having a total of six nozzle holes at the tip, and is provided so as to inject fuel radially from the side on the intake side toward the combustion chamber 6. FIGS. 2A to 2C are diagrams schematically showing the movement of the spray of fuel injected from the injector 15 in the latter half of the compression stroke, and showing the movement of the spray injected from different nozzle holes, respectively. ing. Specifically, the spray X1 shown in FIG. 2 (a) is spray sprayed from one injection hole provided at the highest position in the tip of the injector 15, and the spray shown in FIG. 2 (c). X4 is spray sprayed from one nozzle hole provided at the lowest position, and sprays X2 and X3 shown in FIG. 2B are from two nozzle holes provided at intermediate height positions. The spray is sprayed (in FIG. 2 (c), sprays X2 and X3 are shown in an overlapping manner, but their positions in the depth direction are different).

  As shown in FIG. 2A, after entering the inside of the cavity C, the spray X1 is turned upward along the wall surface W1 on the exhaust side of the cavity C and guided to the vicinity of the spark plug 16. . Further, as shown in FIGS. 2B and 2C, the sprays X2, X3, and X4 do not enter the cavity C and collide with the wall surface on the intake side of the crown surface of the piston 5, and then the raised portion. The direction is changed upward along the inclined wall surface W <b> 2 on the intake side in 5 a and guided to the vicinity of the spark plug 16. Although the remaining two sprays other than these four sprays X1 to X4 are not shown, the remaining two sprays are injected so as to be directed to the outer peripheral portion of the combustion chamber 6 away from the spark plug 16. .

  Thus, in this embodiment, when fuel is injected from the injector 15 in the latter half of the compression stroke, four of the sprays injected from the total of six injection holes (sprays X1 to X4) are all ignited. They are collected in the vicinity of the plug 16. Thereby, a relatively rich air-fuel mixture is formed in the vicinity of the spark plug 16, and so-called fuel stratification is realized. Such fuel stratification is advantageous in improving combustion stability under operating conditions in which combustion stability is likely to be impaired. For example, combustion stability is likely to be impaired under operating conditions that require a retarded ignition timing, such as during a catalyst warm-up operation described later (when a low-temperature catalyst is heated by high-temperature exhaust gas). Therefore, in this embodiment, at least during the catalyst warm-up operation, a part of the fuel to be injected from the injector 15 is injected in the latter half of the compression stroke in order to improve the combustion stability.

  As shown in FIG. 1, the cylinder head 4 is provided with an optical sensor SN3. The optical sensor SN <b> 3 detects a luminous flame that is emission of particulate matter generated in the combustion chamber 6. Here, the particulate matter discharged from the combustion chamber 6 includes soot (soot) resulting from fuel combustion, unburned hydrocarbon (SOF), fuel-derived sulfide (Sulfate), and the like. Although included, according to the result of the research described in item (2) described later, most of the particulate matter generated in the gasoline direct injection engine is soot. Therefore, hereinafter, the particulate matter is simply referred to as soot.

  Each part of the engine configured as described above is centrally controlled by an ECU (engine control unit) 50. As is well known, the ECU 50 is a microprocessor including a CPU, a ROM, a RAM, and the like, and corresponds to a “calculation unit” and an “injection control unit” in the claims.

  Various information is input to the ECU 50 from various sensors provided in the engine. That is, the ECU 50 is electrically connected to the crank angle sensor SN1, the catalyst temperature sensor SN2, and the optical sensor SN3 described above, and based on the input signals from these sensors SN1 to SN3, the crank angle and the engine rotational speed. Various information such as the catalyst temperature and the brightness of the luminous flame (soot emission) is acquired.

  ECU50 controls each part of an engine, performing various calculations etc. based on the information acquired from each said sensors SN1-SN3. Specifically, the ECU 50 is electrically connected to the injector 15, the spark plug 16, the supply pump 21, the throttle valve 30, and the like. Based on the result of the above calculation and the like, control signals are respectively sent to these devices. Is output.

(1-2) Control during catalyst warm-up operation Immediately after the engine is cold-started, the ignition timing is set to activate the low-temperature catalyst at an early stage under conditions where the catalyst temperature is considerably low. The catalyst warm-up control is executed to retard the temperature and raise the exhaust gas temperature (that is, to heat the catalyst with the high-temperature exhaust gas). The details of the catalyst warm-up control will be described below with reference to the flowcharts of FIGS. 3 and 4 and the time chart of FIG.

  When the ignition switch (not shown) is turned on and the processing shown in the flowchart of FIG. 3 is started, the ECU 50 executes control for starting the stopped engine (step S1).

  Next, the ECU 50 determines whether or not the catalyst temperature detected by the catalyst temperature sensor SN2 (the temperature of the catalyst in the catalytic converter 31) is less than a predetermined threshold value Tx (step S2). The threshold value Tx of the catalyst temperature used here is a threshold value for determining the active state of the catalyst. The catalyst temperature being lower than the threshold value Tx indicates that the catalyst needs to be warmed up early. Means inactive.

  If it is determined NO in step S2 and it is confirmed that the catalyst temperature is equal to or higher than the threshold value Tx, that is, if it is confirmed that the catalyst is in an active state or a state close thereto, the catalyst warm-up is performed. Since it is not necessary to execute the control, the routine shifts to a normal operation in which the ignition timing is not retarded.

  On the other hand, when it is determined YES in step S2 and it is confirmed that the catalyst temperature is lower than Tx (the catalyst is inactive), the ECU 50 performs the injector 15 according to the injection pattern 1 shown in FIG. The fuel is injected from (Step S3).

  In the injection pattern 1, as shown in FIG. 5 (a), the first-stage fuel injection F1a is executed at the end of the intake stroke, and the second-stage fuel injection F2a is executed in the second half of the compression stroke. Pattern. In the graph of FIG. 5, since the crank angle (deg. ATDC) when the compression top dead center is 0 ° is adopted on the horizontal axis, the angle on the advance side from the compression top dead center is negative. Expressed numerically. However, in the following description in the text, BTDC indicates that the position is on the advance side of the compression top dead center, and a negative numerical value is not used. For this reason, for example, when referring to an intake bottom dead center of −180 (deg. ATDC) on the graph, it is expressed as BTDC 180 ° CA.

  In the injection pattern 1, as an example, the first-stage fuel injection F1a is started at BTDC 190 ° CA, which is 190 ° earlier than the compression top dead center (TDC), and the 55 ° crank angle is increased from the compression top dead center. The second-stage fuel injection F2a is started at the earlier BTDC 55 ° CA. The illustration here illustrates the injection timing when the engine speed is within a certain range (for example, around 850 rpm). When the engine speed changes, the injection timing is appropriately changed. obtain. The same applies to the injection patterns 2 and 3 (steps S11 and S18) described later.

  Further, in the injection pattern 1, the supply pump 21 is controlled so that the injection pressure (fuel pressure) of the fuel from the injector 15 is lower than in the injection patterns 2 and 3 described later. Specifically, the fuel pressure in the injection pattern 1 is set to about 15 to 20 MPa, for example.

  The ratio (split ratio) between the injection amount by the first-stage fuel injection F1a and the injection amount by the second-stage fuel injection F2a can be set as appropriate. Can be set to

  After executing the fuel injection (step S3) according to the injection pattern 1 as described above, the ECU 50 retards a predetermined amount of delay from the compression top dead center so that the injected fuel starts combustion in the middle of the expansion stroke. The spark plug 16 is operated at the timing set to (step S4).

Next, the ECU 50 detects the luminous flame, which is the soot emission, by the optical sensor SN3 (step S5), and performs an operation for identifying the KL value (KL _θmax ) that serves as an index of the soot amount based on the detected luminous flame. Execute (Step S6).

The KL value is the product of the soot concentration and the thickness of the combustion field (the thickness of the flame with respect to the direction in which the emitted light is detected), which will be described in detail in item (2) described later. Here, the KL value of the entire cylinder at each crank angle is KL _θ , and the peak value is KL _θmax . In step S6, the ECU 50 specifies the brightness of two different wavelength components obtained by splitting the radiation light from the bright flame, obtains the soot temperature (brightness temperature) from each brightness, and further calculates the brightness temperature. KL _θ and KL _θmax are identified by the predetermined calculation used. Although this method is called a two-color method, the two-color method itself is well known, and since the two-color method will be described in item (2) described later, a detailed description of the two-color method is omitted here. .

As described above, in order to obtain the KL values (KL _θ and KL _θmax ), it is necessary to _split the emitted light from the luminous flame into two different wavelength components. For this reason, in this embodiment, the filter for splitting the radiation light input into the optical sensor SN3 from the luminous flame is built in the optical sensor SN3.

After _obtaining KL _θmax as described above, the ECU 50 performs an operation for estimating the number of soot discharged from the engine (hereinafter, this may be abbreviated as PN) based on the obtained KL _θmax. (Step S7). As will be described in detail in item (2) described later, according to the study by the present inventor, PN and KL _θmax are linear (the PN also increases at a substantially constant rate as the value of KL _θmax increases ( There is a linear function relationship (see graph V in FIG. 3). Therefore, the estimated value of PN can be obtained by substituting KL _θmax into a predetermined linear function, for example. Alternatively, an estimated value of PN may be obtained from KL _θmax using a map prepared in advance based on the linear function.

  Next, the ECU 50 determines whether or not the PN estimated in step S7 is less than a predetermined threshold value α (step S8). The threshold value α of PN used here is a threshold value for determining whether or not the spraying condition needs to be improved in order to reduce PN, and that PN is greater than or equal to this threshold value α It means that it is necessary to improve the spraying conditions (thus reducing PN) due to the unexpectedly large number.

  When it is determined YES in Step S8 and it is confirmed that PN is less than the threshold value α, the ECU 50 returns to Step S2 and repeats the subsequent processes.

  On the other hand, when it is determined NO in Step S8 and it is confirmed that PN is equal to or higher than the threshold value α, the ECU 50 proceeds to Step S10 in FIG. 4 to determine whether or not the catalyst temperature is lower than the threshold value Tx. When the determination is YES, that is, when the determination is YES (that is, when the catalyst is still inactive), the fuel injection pattern is switched from the above-described injection pattern 1 to the injection pattern 2 shown in FIG. Then, fuel is injected from the injector 15 according to the injection pattern 2 (step S11).

  In the injection pattern 2, as shown in FIG. 5 (b), the first-stage fuel injection F1b is executed at the end of the intake stroke, and the second-stage fuel injection F2b is executed in the second half of the compression stroke. Pattern. As an example, the first and second fuel injections F1b and F2b are started at BTDC 190/55 ° CA, respectively, as in the case of the injection pattern 1 described above. In addition, although the ratio (split ratio) of the injection amount by the fuel injection F1b and F2b of each stage can be set as appropriate, it can be set to a split ratio of, for example, two equal parts (50:50).

  Further, in the injection pattern 2, the supply pump 21 is controlled so that the injection pressure (fuel pressure) of the fuel from the injector 15 is higher than in the injection pattern 1. Specifically, the fuel pressure in the injection pattern 2 is set to about 20 to 30 MPa, for example.

  After executing the fuel injection by the injection pattern 2 (step S11) as described above, the ECU 50 retards a predetermined amount from the compression top dead center so that the injected fuel starts combustion in the middle of the expansion stroke. The spark plug 16 is operated at the timing set to the side (step S12).

Next, the ECU 50 obtains KL _θmax based on the detected luminous flame in the same manner as in steps S5 and S6 (steps S13 and S14), and estimates PN from KL _{θmax in} the same manner as in step S7. (Step S15).

  Next, the ECU 50 determines whether or not the PN obtained in step S15 is less than the threshold value α (step S16).

  When it is determined as YES in Step S16 and it is confirmed that PN is less than the threshold value α, the ECU 50 returns to Step S10 and repeats the subsequent processes.

  On the other hand, when it is determined NO in Step S16 and it is confirmed that PN is equal to or higher than the threshold value α, the ECU 50 determines whether or not the catalyst temperature is lower than the threshold value Tx (Step S17). If YES (that is, if the catalyst is still inactive), the fuel injection pattern is switched from the above-described injection pattern 2 to the injection pattern 3 shown in FIG. Fuel is injected from the injector 15 (step S18).

  In the injection pattern 3, as shown in FIG. 5C, the first-stage fuel injection F1c is executed at the end of the intake stroke, the second-stage fuel injection F2c is executed in the first half of the compression stroke, and the compression is further performed. This is a three-split injection pattern in which the third-stage fuel injection F3c is executed in the latter half of the stroke. As an example, the first stage fuel injection F1c starts at BTDC 190 ° CA, the second stage fuel injection F2c starts at BTDC 115 ° CA, and the third stage fuel injection F3c starts at BTDC 55 ° CA. In addition, although the ratio (split ratio) of the injection amount by the fuel injection F1c, F2c, F3c of each stage can be set as appropriate, it can be set to a split ratio of, for example, three equal parts (33:33:33).

  Further, in the injection pattern 3, as in the case of the injection pattern 2, the supply pump 21 is controlled so that the injection pressure (fuel pressure) of the fuel from the injector 15 is increased to, for example, about 20 to 30 MPa.

  After executing the fuel injection by the injection pattern 3 (step S18) as described above, the ECU 50 retards a predetermined amount from the compression top dead center so that the injected fuel starts combustion in the middle of the expansion stroke. The spark plug 16 is operated at the timing set to the side (step S19).

(2) Research that is the basis of the present invention As described above, in the embodiment, a luminous flame that is light emission of soot is optically detected to obtain a KL value (more specifically, KL _θmax ), and from the obtained KL value. The fuel injection from the injector 15 is controlled based on the estimated size of PN. The present invention represented by this embodiment has been made on the basis of research conducted by the present inventor in order to easily specify the number of soot (PN). The details of this research are described below.

(2-1) Experimental method (a) Test engine and experimental conditions In the experiment, a single-cylinder visualization engine based on a mass-produced gasoline engine was used. The structure of this visualization engine is shown in FIG. 6, the specifications are shown in Table 1, and the experimental conditions are shown in Table 2. As shown in FIG. 6, the visualization engine uses a cylinder block made of quartz glass (hereinafter referred to as a glass liner), and the inside (combustion chamber) of the engine can be observed from the outside through this glass liner. Yes.

  As explained in the Background Art section, most of the PN discharged from the gasoline direct injection engine is generated during the catalyst warm-up operation immediately after the engine is started. Therefore, in the experiment, in correspondence with the catalyst warm-up operation, the engine is operated so that the engine rotation speed becomes constant at 850 rpm, and the ignition timing is retarded. The reason for retarding the ignition timing is to raise the exhaust gas temperature and activate the catalyst at an early stage. In general, retarding the ignition timing for the purpose of raising the exhaust gas temperature impairs combustion stability. However, in this engine, the stratified combustion concept similar to the embodiment described in item (1) above is sufficient. Combustion stability is ensured. That is, fuel is injected during the compression stroke from an injector having six injection holes, and four sprays injected from the four injection holes are collected around the spark plug so that a long time after compression top dead center is obtained. This makes it possible to supply a stable combustible mixture over a period of time.

(B) Measurement method In measuring PN, the number of particulate matter having a particle diameter of 23 to 1000 nm passing through the exhaust passage is measured using a high-speed response fine particle size distribution analyzer (manufactured by Cambustion, DMS500) that can understand the particle size distribution. did. In addition, it is a thing corresponding to European regulation (Euro6) that the particle size used as measurement object is 23-1000 nm.

  In order to elucidate the phenomenon of PN generation, as shown in FIG. 6, sampling was performed at a position immediately downstream of the exhaust port, and the inside of the combustion chamber was photographed through a glass liner. A high-speed color camera (manufactured by Photon, FASTCAMSA-X) was used for photographing, and a combustion flame and spray scattered light were photographed from the engine front side. An Nd: YLF laser (manufactured by Litron, LDY304) was used to photograph the spray scattered light, and it was converted into sheet light from a cylindrical lens and incident on the bore center in the combustion chamber from the engine exhaust side through a glass liner.

(2-2) Experimental results (a) Cause of particulate generation during catalyst warm-up operation In the experiment, a 6-hole type mass production injector (the position of each nozzle hole is the same as that of the above-described embodiment), The fuel was injected in two stages with a fuel pressure of 18 MPa. Specifically, after the start timing of the first stage fuel injection is fixed at BTDC 190 ° CA (the timing when the crank angle is advanced from the compression top dead center) which is the end of the intake stroke, the second stage fuel is injected. The start timing of the injection was varied in the compression stroke. Further, the ratio of the first and second stage injection amounts was 50:50. Then, the engine was operated under each condition, and the number of particulate matter per cc (PN) and the combustion fluctuation rate, which is the fluctuation rate of the indicated mean effective pressure (IMEP), were measured. The results are shown in FIG. The graph is summarized. In the following, the start timing of the first stage fuel injection is denoted as 1st · SOI, and the start timing of the second stage fuel injection is denoted as 2nd · SOI. The 2nd · SOI time is set to any one of (i) BTDC 70 ° CA, (ii) BTDC 60 ° CA, (iii) BTDC 55 ° CA, (iv) BTDC 50 ° CA, and (v) BTDC 45 ° CA.

  As shown in FIG. 7, as the 2nd · SOI is retarded (that is, in the order of (i) → (ii) → (iii) → (iv) → (v)), the combustion fluctuation rate decreases and PN Is increasing. From this, it can be seen that the PN and the combustion fluctuation rate strongly depend on 2nd · SOI. Note that a decrease in the combustion fluctuation rate means that the combustion stability is improved. That is, as the 2nd · SOI is retarded, a rich air-fuel mixture that easily ignites is more likely to be formed in the vicinity of the spark plug (that is, fuel stratification becomes stronger). Combustion will be realized. On the other hand, the retardation of 2nd · SOI increases the proportion of fuel combusted in a state of insufficient oxygen, which leads to an increase in PN.

  Here, the particulate matter discharged from the gasoline engine is classified into unburned hydrocarbon (SOF), fuel-derived sulfide (Sulfate), etc. in addition to soot caused by combustion, and includes particles of various diameters. It is out. Therefore, FIG. 8 shows the particle size distribution of the particles to be measured in FIG. 7 (that is, the detected particles are classified for each particle size and the number of particles for each particle size category is identified). As shown in FIG. 8, a particle size distribution having a peak in the vicinity of approximately 70 nm was obtained under any of the conditions in which 2nd · SOI was changed. In addition, the number of particles at the peak position increases as the 2nd · SOI is retarded, and generally agrees with the PN trend shown in FIG.

  It is known that soot particles generated by combustion of a gasoline direct injection engine are composed of primary particles having a particle size of 5 to 60 nm and particles having a particle size of 20 to 200 nm in which a plurality of primary particles are aggregated. From this, it is considered that most of the particles (particles having a particle size distribution peak near 70 nm) that are measured in FIG.

  In general, soot discharged from a gasoline direct injection engine is considered to be due to (i) adhesion of fuel spray to the piston wall surface and (ii) richness of the air-fuel mixture. Therefore, in order to consider the cause of the soot generation in this engine, the combustion flame and spray scattered light were photographed, and the results of FIGS. 9 and 10 were obtained. FIG. 9 is a graph showing a graph obtained by measuring an in-cylinder pressure and a heat generation rate with an average of 50 cycles, and a combustion flame image acquired at a plurality of crank angles, and FIG. It is the figure which showed the spray scattered light image by the fuel injection of the 2nd step. In these FIGS. 9 and 10, 2nd · SOI is set to BTDC 55 ° CA in consideration of combustion stability.

  When attention is paid to the flame image shown in FIG. 9, it can be seen that after the propagation flame reaches the piston wall surface, a luminous flame as soot emission appears from the vicinity of the piston wall surface on the intake side (region A). Further, as the piston descends, the luminous flame area expands, and it is observed that the luminous flame is also generated from the vicinity of the cavity of the piston. Further, when attention is paid to the spray image shown in FIG. 10, it is understood that the spray collides with the piston wall surface on the intake side where the luminous flame is confirmed and the vicinity of the cavity. From these facts, PN generation during the catalyst warm-up operation is mainly caused by the so-called pool combustion that generates soot particles by the flame reaching the wall surface before the fuel adhering to the piston wall evaporates and mixes. It is thought to be a cause.

(B) Correlation between luminous flame and PN As shown in FIG. 9, a luminous flame was confirmed from the vicinity of the spray collision position on the piston wall surface. Difficult to do. Therefore, the luminous flame was quantified as a KL value by the two-color method, and the relationship between the KL value and PN was investigated. In order to quantify the luminous flame, two-color method measurement was performed using the method proposed by Hottel and Broughton. This is the same method as described in the embodiment of (1) above, and the soot temperature (luminance temperature) is calculated from the luminances of two different wavelength components obtained by splitting the emitted light from the luminous flame twice. This is an optical method for obtaining the KL value and the flame temperature (the temperature of the flame, which is a reduction in heating of the soot) at the same time, as a result of the calculation process using the luminance temperature. In the two-color measurement, a bright flame is photographed using a high-speed color camera, and the photographed image is introduced into a two-color temperature measurement software (manufactured by Mitsui Optronics, HS-Thermera ver 4.87) for analysis. It was. Since the wavelength of the luminous flame is different from the wavelength of the combustion flame, the luminance of the luminous flame can be specified separately from the flame.

  The KL value is the product of the soot concentration and the thickness of the combustion field (the thickness of the flame with respect to the direction in which the radiated light is detected), as described in the above-described embodiment. This KL value is considered to have a correlation with the soot mass in the cylinder, and is expressed by the following equation (1).

Here, m: mass of soot [kg], V _cyl : volume [m ³ ], L _ls : thickness of combustion field [m]. The theory and calculation method of the two-color method are described in various past documents, and the description thereof is omitted here.

FIG. 11 shows the flame temperature, KL value, and luminance when the two-color method measurement is performed under the same conditions as in FIG. Specifically, the graph of FIG. 11A shows the average flame temperature at each crank angle, and the graph of FIG. 11B shows KL _θ which is the KL value of the entire cylinder at each crank angle. And luminance. The luminance here is the sum of each pixel in the flame image.

As shown in FIG. 11B, the luminance shows a peak in the vicinity of ATDC 90 ° CA delayed by 90 ° from the compression top dead center, but the value at the exhaust valve opening timing (hereinafter referred to as EVO) is a peak value. Compared to, it has decreased significantly. On the other hand, KL _θ shows a peak slightly before EVO, and there is no significant difference between the value at that peak and the value at EVO. As a reason why such a difference appears, the influence of the flame temperature change accompanying the lowering of the piston (expansion of the combustion chamber) can be considered. That is, since the light emission of soot becomes weaker as the flame temperature is lowered, the luminance is easily affected directly by changes in the flame temperature. For this reason, when the crank angle is advanced to EVO and the flame temperature is lowered, the brightness is lowered accordingly, and the difference from the peak value is enlarged. On the other hand, KL _θ is less affected by the change in flame temperature than luminance, so KL _θ maintains a value close to the peak value even when the crank angle advances to EVO. From the above, it can be said that KL _θ more accurately reflects the soot mass in the cylinder than the luminance.

FIG. 12 shows the result of simultaneous measurement of KL _θmax and PN while variously changing 2nd · SOI under the conditions of fuel pressure = 18 MPa, 1st · SOI = BTDC 190 ° CA. Note that KL _θmax is the peak value of KL _θ shown in FIG. Since KL _θ shows a peak in the vicinity of EVO, the peak value (KL _θmax ) was adopted as a representative value of the soot mass in the cylinder.

From FIG. 12, KL _θmax and PN are in a linear (linear function) relationship, and the correlation coefficient R between them is a very high value of about 0.97. That is, it was confirmed that PN increases at a substantially constant rate as the value of KL _θmax increases. From the above results, it was found that PN can be quantitatively predicted using KL _θmax .

(C) PN reduction by improvement of spraying and multi-stage injection From the result of (b) above, the KL value in the cylinder and PN could be linked quantitatively. Moreover, from the visualization result of the above (a), it was presumed that the pool combustion due to the adhesion of the sprayed piston to the wall surface is the main cause of the generation of PN. Therefore, with the aim of reducing the amount of the second-stage injected fuel adhering to the piston wall surface, spraying was improved and multistage injection was performed to reduce PN. Table 3 shows the respective injection conditions.

  Table 3 shows a comparative example in which a mass production injector is used, a comparative example in which an improvement is made to reduce spray penetration with respect to this comparative example, and a specification in which the fuel pressure is increased compared to the first example. Example 2 (18 MPa → 25 MPa) and Example 3 in which multistage injection (2 stages → 3 stages) was further performed with respect to Example 2 were designated as Example 3. The penetration (penetration force) of the spray can be changed by changing at least one of the diameter and axial length of the nozzle hole, but here, the nozzle hole diameter was changed. That is, in Examples 1 to 3, the penetration was reduced (1 → 0.82) by reducing the nozzle hole diameter compared to the comparative example.

  In the rightmost column in Table 3, the numerical value in parentheses written together with the number of injections indicates the start timing of fuel injection at each stage. For example, the numerical value “2 (−190 / −55)” corresponding to the comparative example, the example 1, and the example 2 indicates that the fuel is injected in two stages and the first stage injection start timing (1st This indicates that (SOI) is set to BTDC 190 ° CA, and the second stage injection start timing (2nd · SOI) is set to BTDC 55 ° CA. On the other hand, in the numerical value “3 (−190 / −115 / −55)” corresponding to the third embodiment, the fuel is injected in three stages, and the first stage injection start timing (1st · SOI) is BTDC190. This indicates that the second stage injection start timing (2nd · SOI) is set to BTDC 115 ° CA, and the third stage injection start timing (3rd · SOI) is set to BTDC 55 ° CA, respectively.

  Here, the case where the first embodiment, that is, the fuel pressure is relatively low 18 MPa, and the first stage / second stage injection start timing (1st / 2nd · SOI) is BTDC 190/55 ° CA is described above. This corresponds to the injection pattern 1 (FIG. 5A) in the embodiment. In addition, the second embodiment, that is, the case where the fuel pressure is relatively high, 25 MPa, and the first stage / second stage injection start timing (1st / 2nd · SOI) is BTDC 190/55 ° CA is described above. This corresponds to the injection pattern 2 (FIG. 5B) in the embodiment. Further, in Example 3, that is, the fuel pressure is relatively high 25 MPa, and the first stage / second stage / third stage injection start timing (1st / 2nd / 3rd · SOI) is BTDC 190/115/55 ° CA. This case corresponds to the injection pattern 3 (FIG. 5C) in the above-described embodiment.

FIGS. 13A and 13B show waveforms of KL _θ and heat generation rate under each injection condition, and FIG. 14 shows a distribution image of KL values at ATDC 112 ° CA. Focusing on the waveform of KL _θ (FIG. 13), KL _θmax tended to decrease as the conditions were changed from Comparative Example to Example 1 → Example 2 → Example 3. In particular, in Example 3, it can be seen that KL _θmax is reduced by 71% compared to the comparative example. That is, in the case where the fuel pressure is increased and the number of stages is increased as in the third embodiment, the PN can be reduced by about 70%. Further, as shown in the distribution image of the KL value (FIG. 14), the area of the KL value in the vicinity of the wall surface of the piston is changed as the conditions are changed from Comparative Example to Example 1 → Example 2 → Example 3. It can also be confirmed that is decreasing.

(D) Consideration of the model As described above, it was found that the KL value was reduced by the improvement of the spray and the multistage injection. Next, for the purpose of further optimization, calculation by CFD was performed in accordance with the injection conditions shown in Table 3. The result is shown in FIG. In addition, the vertical axis | shaft of FIG. 15 is the adhesion amount of the fuel to a piston wall surface.

  As shown in FIG. 15, the fuel adhesion amount decreases as the spraying conditions are improved from Comparative Example to Example 1 → Example 2 → Example 3, and the same tendency as the experimental result described in the above (c). Has been reproduced. Therefore, it was found that the PN during the catalyst warm-up operation of the gasoline direct-injection engine can be arranged by the fuel adhesion amount on the piston wall surface and can be examined in detail by CFD.

(E) Result in actual vehicle Based on the results of the combustion experiment using the model engine so far, a test was carried out by mounting an engine to which the spray conditions generally corresponding to Example 3 were applied to an actual vehicle. a) The results shown in (b) were obtained. According to this, it was confirmed that a 66% PN reduction effect can be obtained when traveling in the NEDC test mode compared to before improvement.

(2-3) Summary The conclusions obtained in this study are summarized below.

  (I) The PN measured during the catalyst warm-up operation is a soot mainly caused by combustion, suggesting that it is generated by pool combustion due to fuel adhesion to the piston wall surface.

(Ii) PN during the catalyst warm-up operation has a high correlation with the KL value, and it was confirmed that the PN can be quantitatively predicted from the value of KL _θmax .

(Iii) It was found that KL _θmax during catalyst warm-up operation can be greatly reduced by adopting spray conditions that combine low penetration, high fuel pressure, and multistage injection, and the effect was confirmed even in actual vehicles.

(3) Description of Operational Effects of Embodiments Next, the operational effects of the embodiment shown in FIGS. 1 to 5 will be described based on the results of the above-described research by the inventors of the present application.

In the above embodiment, during the catalyst warm-up operation in which the ignition timing is retarded because the catalyst temperature is low, the luminous flame, which is the soot emission, is detected by the optical sensor SN3 and based on the detected luminous flame. A KL value (KL _θmax ) is obtained by the two-color method, and it is estimated that the larger the KL value is, the more PN is (steps S5 to S7 / S13 to S15). Based on the estimated size of PN, the fuel injection pattern from the injector 15 is variably set between patterns 1 to 3 (FIGS. 5A to 5C). Specifically, when it is confirmed that the PN is equal to or greater than the threshold value α in the state where the injection pattern 1 is selected, the injection pattern is switched from the pattern 1 to the pattern 2 (steps S8 and S11). If it is confirmed that PN is greater than or equal to the threshold value α with 2 selected, the ejection pattern is switched from pattern 2 to pattern 3 (steps S16 and S18).

  On the other hand, according to the result of the research described in the item (2), in Examples 1, 2, and 3 corresponding to the injection patterns 1, 2, and 3, the fuel adheres to the wall surface of the piston 5 in this order. It has been found that the amount can be reduced (FIG. 15). Therefore, as in the above embodiment, when the injection pattern is switched from pattern 1 to pattern 2 or from pattern 2 to pattern 3 in accordance with PN, the piston is moved along with the switching. 5 The amount of fuel adhering to the wall surface can be reduced.

  Thus, in the above embodiment, the fuel injection from the injector 15 is controlled so that the amount of fuel adhering to the wall surface of the piston 5 is reduced when the PN estimated from the KL value is large compared to when the PN is small. Therefore, it is possible to suppress pool combustion that occurs when the flame reaches the fuel in a droplet state attached to the wall surface, and it is possible to suppress generation of soot due to the pool combustion. Thereby, combustion can be appropriately controlled so that soot is not generated excessively, and PN (number of soot) discharged from the engine can be effectively suppressed.

  More specifically, in the above embodiment, the fuel is injected in two times, the final stage of the intake stroke and the latter half of the compression stroke, regardless of which of the injection patterns 1 and 2 is selected. Compared to the injection pattern 1, the fuel injection pressure (fuel pressure) from the injector 15 is set higher (FIGS. 5A and 5B). Thereby, during the catalyst warm-up operation in which the ignition timing by the spark plug 16 is retarded, the injection pattern is switched from the pattern 1 to the pattern 2 as necessary while stratifying the fuel and ensuring good combustion stability. Thereby, atomization of fuel can be accelerated | stimulated and PN can be suppressed.

  That is, in the injection patterns 1 and 2, since the start timing of the second stage fuel injection (F2a, F2b) is delayed until the latter half of the compression stroke, a rich mixture that is easy to ignite immediately before the mixture is ignited. It is formed locally and stratification of fuel is realized. Such fuel stratification is advantageous in improving the situation where the combustion stability is likely to be lowered because the ignition timing is retarded. However, the fact that the second stage fuel injection (F2a, F2b) is delayed until the latter half of the compression stroke means that the amount of fuel adhering to the piston 5 wall surface as droplets increases. For this reason, in some cases, there is a concern that soot is excessively generated due to pool combustion of the attached fuel. On the other hand, in the said embodiment, when many PN is confirmed in the state from which the injection pattern 1 is selected, an injection pattern is switched from the pattern 1 to the pattern 2, and a fuel pressure is raised with the said switching. Therefore, atomization of the fuel is promoted, and the time required for the fuel to vaporize and atomize can be shortened. Thereby, since the fuel adhesion amount to the wall surface mentioned above is reduced, compared with the case of the injection pattern 1, PN can be suppressed.

  On the other hand, in the injection pattern 3, the number of divisions of fuel injection is increased from 2 times to 3 times compared to the injection pattern 2. Specifically, the fuel is injected in three steps, ie, at the end of the intake stroke, the first half of the compression stroke, and the second half of the compression stroke (FIG. 5 (c)). In other words, when the number of divisions is increased, in the injection pattern 3, the injection amount by the final stage (third stage) fuel injection F3c executed at the latest timing in the compression stroke is the injection pattern 2. The injection amount by the fuel injections F1c and F2c before the final stage (first stage and second stage) is conversely reduced, compared to the corresponding injection quantity (injection quantity by the second stage fuel injection F2b). The injection amount corresponding to the pattern 2 (the injection amount by the first stage fuel injection F1b) is increased. Thus, in the above embodiment, the amount of fuel injected at the latest timing during the compression stroke is reduced with the switching from the injection pattern 2 to the injection pattern 3, so that the droplet remains on the wall surface of the piston 5. The amount of adhering fuel can be further reduced, and PN can be effectively suppressed.

  However, when the fuel injection amount at the final stage is reduced as in the injection pattern 3, the stratification of the fuel is weakened, so that the combustion stability is slightly lowered. However, PN suppression is an important issue and should be addressed prior to ensuring combustion stability. On the other hand, in the above-described embodiment, not only the injection pattern 2 that suppresses PN by increasing the fuel pressure (which does not affect the combustion stability) but also suppresses PN by reducing the fuel injection amount at the final stage. Since even injection pattern 3 (which has an effect on combustion stability) is prepared, if PN is not sufficiently reduced even in injection pattern 2, switching to injection pattern 3 further reduces combustion stability. While sacrificing a little, PN can be reliably suppressed. Even if the combustion stability is slightly reduced by switching to the injection pattern 3, this is only during the catalyst warm-up operation, and between the first and third fuel injections F1c and F3c. Since the second-stage fuel injection F2c is added (thus reducing the decrease in combustion stability), it is considered that the sense of discomfort felt by the occupant is minimal.

(4) Modification In the above embodiment, when it is confirmed that the PN estimated from the KL value (KL _θmax ) is less than the threshold value α in a state where any one of the injection patterns 1 to 3 is selected. Maintains the selected current injection pattern. For example, if the estimated PN is sufficiently smaller than the threshold value α, the final stage fuel injection (injection patterns 1 and 2) In this case, there is a possibility that the relationship of PN <α can be maintained even if the timing of the second-stage fuel injections F2a and F2b, and in the case of the injection pattern 3, the timing of the third-stage fuel injection F3c) is further retarded. . Therefore, as a countermeasure for such a case, it is proposed to gradually retard the timing of fuel injection in the final stage while confirming the relationship of PN <α by sensing the KL value. In this way, the timing of fuel injection at the final stage can be set as late as possible while guaranteeing PN <α, so that combustion stability can be further improved. Further, if the combustion stability is improved, the ignition timing can be further retarded, so that the catalyst warm-up performance can be further improved by the retarding.

Further, in the above embodiment, the KL value (KL _θmax ) is obtained by analyzing the luminous flame that is the soot emission by the two-color method, and the PN is estimated from the obtained KL value, and depending on the magnitude of the PN Although the fuel injection is determined, since the KL value and the PN are directly related, the process of estimating the PN from the KL value is omitted, and the injection pattern is determined based on the KL value (that is, The KL value and the injection pattern may be directly linked).

  In the above embodiment, the fuel is injected in two or three times during the catalyst warm-up operation (injection patterns 1 to 3). A pattern for dividing and injecting fuel may be provided.

  Moreover, although the said embodiment demonstrated the example which applied this invention to the gasoline direct injection engine, the engine which can apply this invention is not restricted to this, For example, it is also possible to apply this invention to a diesel engine. .

1 Engine Body 2 Cylinder 5 Piston 15 Injector 31 Catalytic Converter 50 ECU (Calculation Unit, Injection Control Unit)
SN3 optical sensor

Claims (9)

A method for controlling an engine having an injector for injecting fuel into a cylinder,
A first step of optically detecting a luminous flame which is emission of particulate matter present in the cylinder;
A second step of analyzing the luminous flame detected in the first step by a two-color method to obtain a KL value;
A third step of injecting fuel from the injector in such a manner that the number of the particulate matter is suppressed when the KL value obtained in the second step is large compared to when the KL value is small. A characteristic engine control method. The engine control method according to claim 1,
In the third step, the fuel injection from the injector is controlled so that the amount of fuel adhering to the piston wall surface is reduced when the KL value obtained in the second step is large compared to when the KL value is small. An engine control method characterized by: The engine control method according to claim 2,
The injector is controlled to inject fuel at a plurality of timings including a compression stroke during a catalyst warm-up operation where the temperature of the catalyst is low,
In the third step, when the KL value obtained in the second step is large, the fuel injection pressure from the injector is increased compared to when the KL value is small. . The engine control method according to claim 2 or 3,
The injector is controlled to inject fuel at a plurality of timings including a compression stroke during a catalyst warm-up operation where the temperature of the catalyst is low,
In the third step, when the KL value obtained in the second step is large, the amount of fuel injected at the latest timing during the compression stroke is reduced, and compared to the case where the KL value is small. An engine control method characterized by increasing the amount of fuel previously injected. An apparatus for controlling an engine having an injector for injecting fuel into a cylinder,
An optical sensor for optically detecting a luminous flame which is emission of particulate matter present in the cylinder;
An arithmetic unit for analyzing the luminous flame detected by the optical sensor by a two-color method to obtain a KL value;
An injection control unit for injecting fuel from the injector in a manner in which the number of the particulate matter is suppressed when the KL value obtained by the arithmetic unit is large compared to when the KL value is small. The engine control device. A method for detecting the number of particulate matter discharged from an engine,
A first step of optically detecting a luminous flame which is emission of particulate matter present in the cylinder of the engine;
A second step of analyzing the luminous flame detected in the first step by a two-color method to obtain a KL value;
A third step of estimating the number of the particulate matter based on the KL value obtained in the second step,
In the third step, it is estimated that the larger the KL value, the larger the number of particulate matter discharged from the engine. The engine exhaust particle number detection method according to claim 6,
In the second step, the luminance of two different wavelength components obtained by splitting the radiation light from the luminous flame into two spectra is measured, and the KL value is obtained based on the luminance temperature specified from each luminance value. A method for detecting the number of exhaust particles in an engine. The engine exhaust particle number detection method according to claim 7,
In the third step, the number of particulate matter is estimated based on a peak value of the KL value obtained for each crank angle. A device for detecting the number of particulate matter discharged from an engine,
An optical sensor for optically detecting a luminous flame which is emission of particulate matter present in the cylinder of the engine;
An illuminant flame detected by the optical sensor is analyzed by a two-color method to obtain a KL value, and a calculation unit that estimates the number of the particulate matter based on the obtained KL value;
The said calculating part estimates that the number of the said particulate matter discharged | emitted from an engine is so large that the said KL value is large, The exhaust particle number detection apparatus of the engine characterized by the above-mentioned.

Images