JP4946995B2 - Abnormality diagnosis apparatus and abnormality diagnosis method for fuel injection system - Google Patents

Description

  The present invention relates to an abnormality diagnosis device and abnormality diagnosis method for a fuel injection system applied to an internal combustion engine that uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points.

  A catalytic converter for purifying harmful components in exhaust gas is installed in an exhaust system of an internal combustion engine such as a vehicle. One type of such a catalytic converter is a three-way catalytic converter. The three-way catalytic converter oxidizes carbon monoxide and hydrocarbons in exhaust gas and reduces nitrogen oxides to convert them into carbon dioxide, water vapor, and nitrogen.

  The exhaust purification capability of such a three-way catalytic converter greatly depends on the air-fuel ratio of the air-fuel mixture burned in the internal combustion engine. The three-way catalytic converter exhibits the maximum exhaust purification capability when the air-fuel ratio of the air-fuel mixture is the stoichiometric air-fuel ratio. This is because when the air-fuel ratio of the air-fuel mixture is lean and the amount of residual oxygen in the exhaust gas is large, the oxidizing action of carbon monoxide and hydrocarbons becomes active, but the reducing action of nitrogen oxides becomes inactive, and the mixing This is because, when the air-fuel ratio of the gas is rich and the amount of residual oxygen in the exhaust gas is small, the reducing action of nitrogen oxides becomes active, but the oxidizing action of carbon monoxide and hydrocarbons becomes inactive.

  Therefore, in an internal combustion engine equipped with a three-way catalytic converter, air-fuel ratio feedback control is performed in order to maintain the air-fuel ratio of the air-fuel mixture to be burned in the vicinity of the theoretical air-fuel ratio. Air-fuel ratio feedback control is performed by obtaining the deviation between the actual air-fuel ratio and the theoretical air-fuel ratio from the detection result of the oxygen concentration in the exhaust, and performing feedback control on the fuel injection time of the injector so that the deviation is reduced. It has been broken. More specifically, by adjusting the fuel injection time of the injector based on the detection result of the exhaust oxygen concentration so that the oxygen concentration of the exhaust becomes the value at the stoichiometric air-fuel ratio, the air-fuel ratio of the air-fuel mixture is changed to the stoichiometric air-fuel ratio. Control is performed in the vicinity.

  If any abnormality occurs in the fuel injection system under such air-fuel ratio feedback control, the fuel injection characteristics of the injector change greatly, making it difficult to inject and supply the required amount of fuel to the internal combustion engine. As a result, the actual air-fuel ratio may greatly deviate from the stoichiometric air-fuel ratio, and the correction degree of the fuel injection time by the air-fuel ratio feedback control may become extremely large. Therefore, as seen in Patent Document 1, it is determined that there is an abnormality in the fuel injection system when the feedback correction degree of the air-fuel ratio feedback control is excessively large. Diagnosis of deviations from the allowable range of injector injection characteristics has been made.

  A specific example of such an abnormality diagnosis of the fuel injection system will be described. Here, the sum (faf + KG [%]) of the air-fuel ratio feedback correction value faf and the air-fuel ratio learning value KG is used as the index value efafkgd of the correction degree of the fuel injection time in the air-fuel ratio feedback control. Here, it is assumed that the deviation of the index value efafkgd due to the change in the fuel injection rate (fuel injection amount per unit time) of the injector is allowed in the range of “± 35%”. In this case, if the index value efafkgd is within the range of “± 35%”, it is determined as normal, and if it is not within the range, it is determined as abnormal.

  In recent years, on-board internal combustion engines that use a mixed fuel of alcohol and gasoline have been put into practical use. In such an internal combustion engine, various fuels having different alcohol concentrations can be used. Note that the theoretical air-fuel ratio differs between alcohol and gasoline. Specifically, the theoretical air fuel ratio of gasoline is about “14.6”, and the theoretical air fuel ratio of alcohol is about “9”. Therefore, the value of the stoichiometric air-fuel ratio varies depending on the alcohol concentration of the fuel. Therefore, in such an internal combustion engine, in the air-fuel ratio feedback control as described above, the air-fuel ratio that is the control target of the feedback control must be changed according to the alcohol concentration of the fuel. However, the exhaust oxygen concentration when the air-fuel mixture is at the stoichiometric air-fuel ratio is constant regardless of the alcohol concentration of the fuel. Therefore, in the air-fuel ratio feedback control based on the detection result of the exhaust oxygen concentration as described above, the deviation of the air-fuel ratio due to the difference in the alcohol concentration of the fuel is automatically corrected. Even if the air-fuel ratio changes, the air-fuel ratio of the air-fuel mixture is controlled in the vicinity of the stoichiometric air-fuel ratio.

On the other hand, while the air-fuel ratio feedback control is stopped, the air-fuel ratio deviation due to the difference in the alcohol concentration of the fuel is not automatically corrected. And drivability deteriorates. Therefore, as seen in Patent Document 2, conventionally, a learning value (alcohol concentration learning value EG) for correcting the deviation of the air-fuel ratio due to the difference in the alcohol concentration of the fuel is used as the air-fuel ratio feedback correction value faf before and after refueling. We ask from change of. Then, by correcting the fuel injection time according to the alcohol concentration learning value EG, the air-fuel ratio can be appropriately controlled even when the air-fuel ratio feedback control is stopped. Further, the accuracy and responsiveness of the air-fuel ratio feedback control can be improved by incorporating the alcohol concentration learning value EG into the calculation of the fuel injection time even during the execution of the air-fuel ratio feedback control.
JP 2007-127076 A JP 2008-51063 A

  Incidentally, even in an internal combustion engine that uses a mixed fuel and learns the alcohol concentration in the mixed fuel as shown in Patent Document 2, the correction degree of the fuel injection time at the time of air-fuel ratio feedback control as described above. It is conceivable to adopt an abnormality diagnosis of the fuel injection system based on the above. However, in such a case, the following problems occur.

  Even in such a case, if the alcohol concentration learning value is properly learned, abnormality diagnosis of the fuel injection system can be performed without any particular problem. However, in an internal combustion engine that performs learning of the alcohol concentration as described above, it is necessary to consider the possibility that the alcohol concentration learning value EG has been mislearned. An abnormality diagnosis of the injection system may not be performed. That is, in the internal combustion engine as described above, the deviation of the index value efafkgd [%] (= faf + KG) occurs not only due to a change in the fuel injection rate of the injector but also as a result of erroneous learning of the alcohol concentration learning value. Therefore, even if there is no problem in the fuel injection rate of the injector, the correction degree of the fuel injection time in the air-fuel ratio feedback control may become excessive due to the erroneous learning of the alcohol concentration learning value. There is a risk of erroneously determining that there is an abnormality.

  In order to avoid such erroneous determination, it is necessary to relax the abnormality determination condition in the diagnosis by the amount of deviation of the index value efafkgd due to erroneous learning of the alcohol concentration learning value EG. That is, if the alcohol concentration learning value EG is mislearned to a value smaller than the original value, the index value efafkgd is larger than when the alcohol concentration learning value EG is properly learned. In addition, if the alcohol concentration learning value EG is erroneously learned to a value larger than the original value, the index value efafkgd is smaller than when it is properly learned. Therefore, even if the deviation of the index value efafkgd due to the change in the fuel injection rate (fuel injection amount per unit time) of the injector is within the allowable range (in the above example, the range of “± 35%”). If there is an erroneous learning of the alcohol concentration learning value EG, the value of the index value efafkgd (= faf + KG [%]) may be lower than the lower limit value of the allowable range or higher than the upper limit value thereof. Therefore, the upper limit value and the lower limit value of the range of values that can be taken by the index value efafkgd in the state where there is no abnormality in the fuel injection system in consideration of the influence of erroneous learning of the alcohol concentration learning value EG, are determined as normal / abnormal determination. Unless it is set to the threshold value, the determination that there is an abnormality cannot be determined. In such a case, the range of the index value efafkgd that is regarded as normal is greatly expanded as compared with the case where the erroneous learning of the alcohol concentration learning value EG is not considered.

  In this way, if erroneous learning of the alcohol concentration learning value EG is taken into consideration and the abnormality determination conditions are eased accordingly, erroneous determination due to erroneous learning of the alcohol concentration learning value EG can surely be avoided. However, if the abnormality determination condition is relaxed in this way, the allowable range of the fuel injection rate of the injector will be greatly expanded. As long as it is not determined to be abnormal, the emission of the internal combustion engine must be guaranteed, and the emission guarantee range for deviation from the rating of the fuel injection rate of the injector will be greatly expanded. And if the guaranteed range of emissions is expanded, it becomes more difficult to adapt the control amounts of the internal combustion engine. Such a problem is not limited to an internal combustion engine that uses a mixed fuel of alcohol and gasoline, but can also occur in an internal combustion engine that uses a mixed fuel of two types of fuels having different theoretical air-fuel ratios. It has become.

  The present invention has been made in view of such circumstances, and the problem to be solved is to expand the emission guarantee range in an internal combustion engine using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points. An object of the present invention is to provide a fuel injection system abnormality diagnosis device and abnormality diagnosis method capable of accurately performing abnormality diagnosis of a fuel injection system while suppressing the above.

Hereinafter, means for solving the above-described problems and the effects thereof will be described.
In order to solve the above-mentioned problem, the invention described in claim 1 uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and determines the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration. This is applied to an internal combustion engine that performs feedback control of the fuel injection time in order to reduce the deviation, and learns the correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the feedback control as the fuel concentration learning value. In the fuel injection system abnormality diagnosis device for diagnosing abnormality of the fuel injection system of the internal combustion engine based on the correction degree of the fuel injection time in the feedback control, calculation of the index value of the correction degree in the diagnosis the fuel concentration learning value by reflecting the comprises a determination mode changing means for changing in accordance with the normal in the diagnosis, the abnormality determination aspect to the fuel concentration learning value, said The engine, the steady-state deviations of the fuel injection time in the feedback control is learned as an air-fuel ratio learned value, the index value is made is calculated using the product of the fuel concentration learned value and its air-fuel ratio learned value That is the gist.

In order to solve the above-mentioned problem, the invention according to claim 5 as a method for diagnosing abnormality of the fuel injection system uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points. Based on the feedback control of the fuel injection time so as to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio, the correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the feedback control is set as the fuel concentration. is applied to an internal combustion engine to learn a learning value, a method of diagnosis based abnormality of the fuel injection system of the internal combustion engine to the correct degree of the fuel injection time in the feedback control, the correction in the diagnostic by reflecting the fuel concentration learning value in the calculation of the index value of the degree, normal in the diagnosis, as well as changed according to the fuel concentration learning value determined manner in which an anomaly, the In combustion engine, the steady-state deviations of the fuel injection time in the feedback control is learned as an air-fuel ratio learned value, that calculates the index value using the product of the fuel concentration learned value and its air-fuel ratio learned value This is the gist.

  An internal combustion engine to which the abnormality diagnosis device and abnormality diagnosis method are applied uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points. In the internal combustion engine, feedback control of the fuel injection time is performed to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, and the mixing of the two types of fuel in the feedback control is performed. The correction value of the fuel injection time corresponding to the difference in the ratio is learned as the fuel concentration learning value. The fuel concentration learning value here is set to a larger value as the concentration of the fuel having the smaller theoretical air-fuel ratio point (having a richer theoretical air-fuel ratio point) in the mixed fuel as described above is higher. Will be.

  If an abnormality occurs in the fuel injection system of the internal combustion engine and the fuel injection rate of the injector changes greatly, the time required for injection of the required amount of fuel will change significantly, and the correction degree of the fuel injection time in the air-fuel ratio feedback control will be remarkably high. Increase. Therefore, in the abnormality diagnosis device and abnormality diagnosis method, the presence or absence of abnormality in the fuel injection system is diagnosed based on the correction degree of the fuel injection time in the air-fuel ratio feedback control as described above. When learning the fuel concentration learning value as described above, the correction degree of the fuel injection time in the air-fuel ratio feedback control may greatly increase depending on the result of the erroneous learning. Accordingly, in consideration of the effect of mislearning of the fuel concentration learning value, the determination result that there is an abnormality cannot be determined unless the abnormality determination conditions in the fuel injection system diagnosis are relaxed. The range of deviation of the allowable fuel injection rate from the rating is expanded, and as a result, the emission guarantee range is expanded.

  However, the tendency of deviation of the fuel injection time correction value of the air-fuel ratio feedback control due to the erroneous learning of the fuel concentration learning value is different depending on the set value of the fuel concentration learning value at that time. For example, if the fuel concentration learning value is set to the upper limit of the possible range, the fuel concentration learning value may be overestimated, that is, the fuel concentration learning value may be mislearned as a value larger than the original value. But there is no possibility of the reverse. For this reason, in this case, the deviation of the fuel injection time correction value of the air-fuel ratio feedback control due to the erroneous learning of the fuel concentration learning value occurs only on the rich side, that is, on the side where the fuel injection time is corrected to decrease. become. If the fuel concentration learning value is set to the lower limit of the possible range, there is a possibility that the fuel concentration learning value is underestimated, that is, the fuel concentration learning value is erroneously learned as a value smaller than the original value. Even so, there is no possibility of the reverse. Therefore, in this case, the deviation of the fuel injection time correction value of the air-fuel ratio feedback control due to the erroneous learning of the fuel concentration learning value occurs only on the lean side, that is, on the side where the fuel injection time is increased and corrected. become. When the fuel concentration learning value is set to an intermediate value between the upper limit value and the lower limit value of the possible range, the closer the value is to the upper limit value, the more the fuel injection time correction value that can be caused by the erroneous learning. The range of deviation increases on the rich side and decreases on the lean side. Thus, the range of deviation of the fuel injection time correction value that can be caused by the erroneous learning can be limited by the set value of the fuel concentration learning value at that time.

  Therefore, even if the abnormality determination condition in the fuel injection system diagnosis is relaxed in consideration of the effect of mislearning of the fuel concentration learning value, the mitigation situation is limited to some extent by the set value of the fuel concentration learning value at that time. It becomes possible to do. In other words, the closer the fuel concentration learning value at that time is to the upper limit of the range that can be taken, the larger the range of relaxation of the abnormality determination condition is on the rich side and the smaller is on the lean side, the fuel concentration learning While avoiding misjudgment in the fuel injection system abnormality diagnosis due to mislearning of values, it is possible to keep the emission guarantee range small. On the other hand, in the above abnormality diagnosis device and abnormality diagnosis method, the normal / abnormal determination mode in the abnormality diagnosis of the fuel injection system based on the correction degree of the fuel injection time in the air-fuel ratio feedback control is changed according to the fuel concentration learning value. By changing the abnormality determination mode as described above, it is possible to accurately perform abnormality diagnosis of the fuel injection system while suppressing the expansion of the emission guarantee range.

In addition, the change of the normal / abnormal determination mode in the abnormality diagnosis of the fuel injection system according to the fuel concentration learning value as described above is performed by changing the index value of the correction degree of the fuel injection time in the air-fuel ratio feedback control in the diagnosis. The calculation can be performed by reflecting the fuel concentration learning value. In this case, if the index value is calculated so that the larger the set value of the fuel concentration learning value at that time is, the index value is generally corrected to the lean side by that value, the fuel concentration learning value becomes the upper limit value. It is possible to increase the range of relaxation of the abnormality determination condition on the rich side and decrease on the lean side as the value is closer to.

Further, if the air-fuel ratio learned value and the fuel concentration learned value as described above are properly learned together, the air-fuel ratio learning value of the value of the air-fuel ratio feedback control due to a change in the fuel injection rate fuel injection time It can be an effective index value of the correction degree. In the abnormality diagnosing device and the abnormality diagnosing method, such an air-fuel ratio learned value is not used as it is, but the index value is calculated by using the product of this and the fuel concentration learned value. The index value calculated in this way increases as the fuel concentration learning value increases. Therefore, the larger the set value of the fuel concentration learning value at that time, the more the index value can be calculated so that the index value is generally corrected to the lean side by that value.

The invention according to claim 2 uses a fuel mixture of two kinds of fuels having different theoretical air-fuel ratio points, and reduces the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration. Applied to an internal combustion engine that performs feedback control of injection time and learns as a fuel concentration learning value a correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the feedback control. In the fuel injection system abnormality diagnosis device for diagnosing an abnormality in the injection system based on the correction degree of the fuel injection time in the feedback control, the diagnosis is performed by the feedback control excluding the amount depending on the fuel concentration learning value. wherein the correction value of the fuel injection time is performed using as an index value of the correction degree, normally, said threshold value of the index value as the abnormality judgment reference fuel in the previous SL diagnosis By changing in accordance with the concentration learned value, and further comprising a judging mode change means for the diagnosis normal, the change in response to abnormality determination aspect to the fuel concentration learning value as its gist. The invention according to claim 6 uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and reduces the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration. The present invention is applied to an internal combustion engine that performs feedback control of the fuel injection time and learns a fuel injection time correction value corresponding to the difference in the fuel mixing ratio in the feedback control as a fuel concentration learning value. A method for diagnosing abnormality of a fuel injection system based on a correction degree of the fuel injection time in the feedback control, wherein the fuel injection time is corrected in the feedback control excluding the amount depending on the fuel concentration learning value The diagnosis is performed using a value as an index value of the correction degree, and a threshold value of the index value serving as a normal / abnormal determination criterion in the diagnosis is set as the fuel concentration learning value. It is changed according, normal in the diagnosis, as its gist to change according to the fuel concentration learning value determined manner in which an anomaly.

An internal combustion engine to which the abnormality diagnosis device and abnormality diagnosis method are applied uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points. In the internal combustion engine, feedback control of the fuel injection time is performed to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, and the mixing of the two types of fuel in the feedback control is performed. The correction value of the fuel injection time corresponding to the difference in the ratio is learned as the fuel concentration learning value. The fuel concentration learning value here is set to a larger value as the concentration of the fuel having the smaller theoretical air-fuel ratio point (having a richer theoretical air-fuel ratio point) in the mixed fuel as described above is higher. Will be.
If an abnormality occurs in the fuel injection system of the internal combustion engine and the fuel injection rate of the injector changes greatly, the time required for injection of the required amount of fuel will change significantly, and the correction degree of the fuel injection time in the air-fuel ratio feedback control will be remarkably high. Increase. Therefore, in the abnormality diagnosis device and abnormality diagnosis method, the presence or absence of abnormality in the fuel injection system is diagnosed based on the correction degree of the fuel injection time in the air-fuel ratio feedback control as described above. When learning the fuel concentration learning value as described above, the correction degree of the fuel injection time in the air-fuel ratio feedback control may greatly increase depending on the result of the erroneous learning. Accordingly, in consideration of the effect of mislearning of the fuel concentration learning value, the determination result that there is an abnormality cannot be determined unless the abnormality determination conditions in the fuel injection system diagnosis are relaxed. The range of deviation of the allowable fuel injection rate from the rating is expanded, and as a result, the emission guarantee range is expanded.
However, the tendency of deviation of the fuel injection time correction value of the air-fuel ratio feedback control due to the erroneous learning of the fuel concentration learning value is different depending on the set value of the fuel concentration learning value at that time. For example, if the fuel concentration learning value is set to the upper limit of the possible range, the fuel concentration learning value may be overestimated, that is, the fuel concentration learning value may be mislearned as a value larger than the original value. But there is no possibility of the reverse. For this reason, in this case, the deviation of the fuel injection time correction value of the air-fuel ratio feedback control due to the erroneous learning of the fuel concentration learning value occurs only on the rich side, that is, on the side where the fuel injection time is corrected to decrease. become. If the fuel concentration learning value is set to the lower limit of the possible range, there is a possibility that the fuel concentration learning value is underestimated, that is, the fuel concentration learning value is erroneously learned as a value smaller than the original value. Even so, there is no possibility of the reverse. Therefore, in this case, the deviation of the fuel injection time correction value of the air-fuel ratio feedback control due to the erroneous learning of the fuel concentration learning value occurs only on the lean side, that is, on the side where the fuel injection time is increased and corrected. become. When the fuel concentration learning value is set to an intermediate value between the upper limit value and the lower limit value of the possible range, the closer the value is to the upper limit value, the more the fuel injection time correction value that can be caused by the erroneous learning. The range of deviation increases on the rich side and decreases on the lean side. Thus, the range of deviation of the fuel injection time correction value that can be caused by the erroneous learning can be limited by the set value of the fuel concentration learning value at that time.
Therefore, even if the abnormality determination condition in the fuel injection system diagnosis is relaxed in consideration of the effect of mislearning of the fuel concentration learning value, the mitigation situation is limited to some extent by the set value of the fuel concentration learning value at that time. It becomes possible to do. In other words, the closer the fuel concentration learning value at that time is to the upper limit of the range that can be taken, the larger the range of relaxation of the abnormality determination condition is on the rich side and the smaller is on the lean side, the fuel concentration learning While avoiding misjudgment in the fuel injection system abnormality diagnosis due to mislearning of values, it is possible to keep the emission guarantee range small. On the other hand, in the above abnormality diagnosis device and abnormality diagnosis method, the normal / abnormal determination mode in the abnormality diagnosis of the fuel injection system based on the correction degree of the fuel injection time in the air-fuel ratio feedback control is changed according to the fuel concentration learning value. By changing the abnormality determination mode as described above, it is possible to accurately perform abnormality diagnosis of the fuel injection system while suppressing the expansion of the emission guarantee range.
When the correction value of the fuel injection time in the air-fuel ratio feedback control excluding the amount depending on the fuel concentration learning value is used as an index value of the correction degree in the abnormality diagnosis of the fuel injection system, the fuel concentration learning as described above is used. Changes in the normal / abnormal determination mode in the abnormality diagnosis of the fuel injection system according to the value are obtained by learning the fuel concentration learning based on the threshold value (normal / abnormal determination value) of the index value related to the normal / abnormal determination in the diagnosis. This can be done by changing according to the value. Specifically, the larger the set value of the fuel concentration learning value at that time, the smaller the threshold values on the rich side and the lean side, that is, by setting the rich side, the fuel concentration learning value is closer to the upper limit value. As a result, the range of relaxation of the abnormality determination condition can be increased on the rich side and decreased on the lean side.

In order to solve the above-mentioned problem, the invention described in claim 3 uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and determines the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration. This is applied to an internal combustion engine that performs feedback control of the fuel injection time in order to reduce the deviation, and learns the correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the feedback control as the fuel concentration learning value. In the fuel injection system abnormality diagnosis device for diagnosing abnormality of the fuel injection system of the internal combustion engine based on the correction degree of the fuel injection time in the feedback control, a steady deviation of the fuel injection time in the feedback control As an air-fuel ratio learning value, and the air-fuel ratio learning value is set to “KG [%]”, and the fuel injection time feedback corresponding to the deviation is performed. When the correction value is “faf [%]” and the fuel concentration learning value is “EG [%]”, the value “fafkgalcd” obtained by the relational expression “fafkgalcd = faf + EG × KG” is within the specified normal range. The gist is to perform the diagnosis by determining that there is an abnormality in the fuel injection system when deviating.

In order to solve the above-mentioned problem, the invention according to claim 7 as a method for diagnosing abnormality of the fuel injection system uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and produces an exhaust oxygen concentration detection result. Based on the feedback control of the fuel injection time so as to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio, the correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the feedback control is set as the fuel concentration. A method applied to an internal combustion engine that learns as a learning value and diagnoses an abnormality in a fuel injection system of the internal combustion engine based on a correction degree of the fuel injection time in the feedback control, the fuel in the feedback control The steady deviation of the injection time is learned as the air-fuel ratio learning value, and the air-fuel ratio learning value is set to “KG [%]”, and the fuel injection time corresponding to the deviation is Assuming that the feedback correction value is “faf [%]” and the fuel concentration learning value is “EG [%]”, the value “fafkgald” obtained by the relational expression “fakgalcd = faf + EG × KG” is defined as normal. The gist is to perform the diagnosis by determining that the fuel injection system is abnormal when it deviates from the range.

  As described above, by reflecting the fuel concentration learning value in the calculation of the index value of the correction degree of the fuel injection time in the air-fuel ratio feedback control in the abnormality diagnosis of the fuel injection system, while suppressing the expansion of the emission guarantee range, An abnormality diagnosis of the fuel injection system can be accurately performed. A specific example of the reflection of the fuel concentration learned value in the index value of the correction degree at this time can be performed by obtaining the index value fakgalkd by the above relational expression.

In order to solve the above-mentioned problem, the invention described in claim 4 uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and determines the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration. This is applied to an internal combustion engine that performs feedback control of the fuel injection time in order to reduce the deviation, and learns the correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the feedback control as the fuel concentration learning value. In the fuel injection system abnormality diagnosis device for diagnosing abnormality of the fuel injection system of the internal combustion engine based on the correction degree of the fuel injection time in the feedback control, a steady deviation of the fuel injection time in the feedback control As an air-fuel ratio learning value, and the air-fuel ratio learning value is set to “KG [%]”, and the fuel injection time feedback corresponding to the deviation is performed. The correction value is “faf [%]”, the coefficient value of the fuel concentration learning value is “kkinj [times]”, the division value of the cylinder inflow air mass by the reference air-fuel ratio is “K (KL) [g]”, fuel injection Assuming that the reciprocal of the rated value of the rate is “α [sec / g]”, the fuel injection is a value “TAU” obtained by the relational expression “TAU = α × K (KL) × (1 + FAF / 100 + KG / 100) × kkinj”. When the time “fakgalcd = faf + {kkinj × (1 + KG / 100) −1} × 100” is obtained, the fuel injection system is abnormal when a value “fakgalcd” obtained from the relational expression deviates from a specified normal range. The gist is that the diagnosis is made by determining the presence.

In order to solve the above-mentioned problem, the invention according to claim 8 as a method for diagnosing abnormality of the fuel injection system uses a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points. Based on the feedback control of the fuel injection time so as to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio, the correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the feedback control is set as the fuel concentration. A method applied to an internal combustion engine that learns as a learning value and diagnoses an abnormality in a fuel injection system of the internal combustion engine based on a correction degree of the fuel injection time in the feedback control, the fuel in the feedback control The steady deviation of the injection time is learned as the air-fuel ratio learning value, and the air-fuel ratio learning value is set to “KG [%]”, and the fuel injection time corresponding to the deviation is The feedback correction value is “faf [%]”, the coefficient value of the fuel concentration learning value is “kkinj [times]”, and the divided value of the Linder inflow air mass by the reference air-fuel ratio is “K (KL) [g]”, A value “TAU” obtained by a relational expression “TAU = α × K (KL) × (1 + FAF / 100 + KG / 100) × kkinj”, where the reciprocal of the rated value of the fuel injection rate is “α [sec / g]”. When the fuel injection time is set, when the value “fakgalcd” obtained from the relational expression “” fakgalcd = faf + {kkinj × (1 + KG / 100) −1} × 100 ”deviates from the specified normal range, The gist of the present invention is to determine that the fuel injection system is abnormal and perform the diagnosis.

  When the fuel concentration learning value is given as a magnification, the fuel injection system is diagnosed by using the index value fafkgalc obtained by the above relational expression, thereby suppressing the expansion of the emission guarantee range and the fuel injection system. This makes it possible to accurately diagnose abnormalities. The reference air-fuel ratio is a constant set as a reference value of the theoretical air-fuel ratio of the mixed fuel that changes in accordance with the fuel mixture ratio, and an arbitrary value can be set. The theoretical air-fuel ratio of one of the seed fuels is set to that value.

  Hereinafter, an embodiment embodying an abnormality diagnosis device and abnormality diagnosis method for a fuel injection system according to the present invention will be described in detail with reference to FIGS.

  FIG. 1 schematically shows a configuration of an internal combustion engine to which the abnormality diagnosis device and abnormality diagnosis method for a fuel injection system and an abnormality diagnosis method of the present embodiment are applied. This internal combustion engine is configured as an in-vehicle internal combustion engine that uses a mixed fuel of alcohol and gasoline. In this internal combustion engine, it is possible to use a mixed fuel of alcohol and gasoline having an alcohol concentration in the range of “0%” to “85%”.

  As shown in the figure, the internal combustion engine includes an intake passage 10, a combustion chamber 11 and an exhaust passage 12. In the intake passage 10, an air cleaner 13 for purifying the intake air, an air flow meter 14 for detecting the intake air amount, and a throttle valve 15 for adjusting the intake air amount are arranged in this order from the upstream side. The intake passage 10 is branched for each cylinder of the internal combustion engine in the intake manifold 16a downstream of the throttle valve 15, and is connected to the intake port 16 of each cylinder formed in the cylinder head of the internal combustion engine. Each cylinder is provided with an injector 17 for injecting and supplying fuel during intake. The injector 17 of each cylinder is pumped from a fuel tank 19 by a fuel pump 18 and further delivered to a delivery pipe 20. The fuel once stored is distributed and supplied.

  The intake port 16 is connected to the combustion chamber 11 via an intake valve 21 and is opened to the combustion chamber 11 by the intake valve 21 at a necessary time. The combustion chamber 11 is provided with a spark plug 22 that sparks and ignites an air-fuel mixture introduced into the combustion chamber 11. The combustion chamber 11 is connected to an exhaust port 24 via an exhaust valve 23 that is opened when necessary.

  The exhaust passage 12 of the internal combustion engine is merged into one at an exhaust manifold 25 connected downstream of the exhaust port 24 of each cylinder. The exhaust passage 12 is opened to the atmosphere through an oxygen concentration sensor 26 that detects the oxygen concentration in the exhaust and a three-way catalytic converter 27 that purifies the exhaust.

  Such control of the internal combustion engine is performed by the electronic control unit 28. The electronic control unit 28 temporarily stores a central processing unit (CPU) that performs various arithmetic processes related to engine control, a read-only memory (ROM) that stores programs and data for engine control, and CPU calculation results. A random access memory (RAM), and an input / output port (I / O) for exchanging signals with the outside. The input port of the electronic control unit 28 includes the air flow meter 14, the oxygen concentration sensor 26, an accelerator sensor 29 for detecting the accelerator operation amount, a throttle sensor 30 for detecting the throttle opening, and an NE for detecting the engine speed. Various sensors such as a sensor 31 and a remaining amount sensor 32 for detecting the remaining amount of fuel in the fuel tank 19 are connected. The electronic control unit 28 performs engine control by driving the throttle valve 15, the injector 17, the spark plug 22, and the like based on the detection results of these sensors.

  Next, fuel injection control in such an internal combustion engine will be described. In this internal combustion engine, the electronic control unit 28 performs feedback control of the fuel injection time, so-called air-fuel ratio feedback control, to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration. ing. During this air-fuel ratio feedback control, the electronic control unit 28 calculates the fuel injection time TAU [sec] of the injector 17 based on the following equation (1).

ここで「α」は、燃料噴射量［ｇ］を燃料噴射時間［ｓｅｃ］に換算するための係数［ｓｅｃ／ｇ］であり、その値には、インジェクタ１７の燃料噴射率［ｇ／ｓｅｃ］の定格値の逆数が設定されている。また上式（１）の「Ｋ（ＫＬ）」は、ベース燃料噴射量［ｇ］であり、その値は、シリンダ流入空気質量ＫＬを基準空燃比にて除算して求められている。ここでシリンダ流入空気質量ＫＬとは、気筒内に流入する空気の質量であり、その値は、エアフローメータ１４による吸入空気量の検出値、ＮＥセンサ３１による機関回転速度の検出値から求められている。また基準空燃比とは、燃料の混合比率に応じて変化する混合燃料の理論空燃比の基準値として設定される定数であり、ここではガソリンの理論空燃比（＝約１４．６）がその値に設定されている。なお上記係数αとベース燃料噴射量Ｋ（ＫＬ）との積は、アルコール濃度「０％」の混合燃料、すなわちガソリンの使用時であって、インジェクタ１７の燃料噴射率が定格通りであるときの、理論空燃比を得るために必要な燃料噴射時間の理論値となっている。

Here, “α” is a coefficient [sec / g] for converting the fuel injection amount [g] into the fuel injection time [sec], and the value is the fuel injection rate [g / sec] of the injector 17. The inverse of the rated value is set. Also, “K (KL)” in the above equation (1) is the base fuel injection amount [g], and the value is obtained by dividing the cylinder inflow air mass KL by the reference air-fuel ratio. Here, the cylinder inflow air mass KL is the mass of air flowing into the cylinder, and the value is obtained from the detected value of the intake air amount by the air flow meter 14 and the detected value of the engine speed by the NE sensor 31. Yes. The reference air-fuel ratio is a constant set as a reference value of the stoichiometric air-fuel ratio of the mixed fuel that changes in accordance with the fuel mixture ratio. Here, the stoichiometric air-fuel ratio (= about 14.6) of gasoline is the value. Is set to Note that the product of the coefficient α and the base fuel injection amount K (KL) is obtained when a mixed fuel having an alcohol concentration of “0%”, that is, gasoline, is used and the fuel injection rate of the injector 17 is as rated. This is the theoretical value of the fuel injection time necessary to obtain the theoretical air-fuel ratio.

  On the other hand, “faf” in the above equation (1) is the air-fuel ratio feedback correction value [%], and the value is set according to the detection result of the exhaust oxygen concentration by the oxygen concentration sensor 26. That is, when calculating the air-fuel ratio feedback correction value faf, first, the deviation between the detected value of the exhaust oxygen concentration by the oxygen concentration sensor 26 and the value at the theoretical air-fuel ratio is obtained. Here, the value of the stoichiometric air-fuel ratio itself changes according to the alcohol concentration of the mixed fuel. However, the exhaust oxygen concentration, that is, the concentration of oxygen surplus in combustion, is always constant at the time of combustion at the stoichiometric air-fuel ratio, whatever the value. Therefore, the degree of deviation of the actual air-fuel ratio from the stoichiometric air-fuel ratio can be determined from the exhaust oxygen concentration without depending on the theoretical air-fuel ratio of the mixed fuel at that time. Then, the value of the air-fuel ratio feedback correction value faf is decreased by a constant value every control cycle if the exhaust oxygen concentration at that time is thinner than the theoretical air-fuel ratio and the actual air-fuel ratio is richer than the stoichiometric air-fuel ratio. If the exhaust oxygen concentration at that time is higher than that at the stoichiometric air-fuel ratio and the actual air-fuel ratio is leaner than the stoichiometric air-fuel ratio, the engine is operated in such a manner that it is increased by a constant value every control cycle. Yes.

  Also, “KG” in the above equation (1) is an air-fuel ratio learning value [%], and this value is obtained by learning the steady deviation of the fuel injection time TAU in the air-fuel ratio feedback control. The air-fuel ratio learning value KG is such that the changes in vehicle speed, engine speed, and intake air amount are small, the operating conditions of the internal combustion engine are stable, and the absolute value of the air-fuel ratio feedback correction value faf is larger than a certain level. It is updated when a learning condition such as that the state has continued for a certain time or longer is satisfied. The air-fuel ratio learning value KG is updated by adding the average value fafav of the air-fuel ratio feedback correction value faf at that time to the value before the update.

  Further, “kkinj” in the above equation (1) is a coefficient value [times] of the alcohol concentration learning value, and this value is fuel injection corresponding to the difference in the fuel mixing ratio (alcohol concentration) in the air-fuel ratio feedback control. The time correction value is learned. The coefficient value kkinj of the alcohol concentration learning value is updated on the condition that the fuel supply to the fuel tank 19 is confirmed. That is, when the remaining amount of fuel in the fuel tank 19 detected by the remaining amount sensor 32 increases by a certain value or more and it is confirmed that refueling is performed, the electronic control unit 28 temporarily stops updating the air-fuel ratio learning value KG. To do. Here, assuming that the air-fuel ratio learned value KG has been properly learned before refueling, all changes in the air-fuel ratio feedback correction value faf after refueling in a state where the update of the air-fuel ratio learned value KG has been stopped are all fuel before and after refueling. It can be considered that it is caused by the change in alcohol concentration. Therefore, after the refueling is confirmed, the electronic control unit 28 increases or decreases the coefficient value kkinj of the alcohol concentration learning value in conjunction with the increase or decrease of the air-fuel ratio feedback correction value faf during the period until the change of the air-fuel ratio feedback correction value faf converges. . As a result, the change in the air-fuel ratio feedback correction value faf resulting from the change in the alcohol concentration of the fuel is accurately reflected in the coefficient value kkinj of the alcohol concentration learning value. In this internal combustion engine, the appropriate value of the coefficient value kkinj is “1.0” when the alcohol concentration of the fuel is “0%”, and “1” when the alcohol concentration of the fuel is the maximum “85%”. .383 ". That is, the coefficient value kkinj takes a value from the minimum value “1.0” to the maximum value “1.383”.

  Meanwhile, the electronic control unit 28 performs abnormality diagnosis of the fuel injection system during engine operation as part of engine control. This abnormality diagnosis is for diagnosing whether or not there is a deviation from the allowable range of the fuel injection rate due to the malfunction of the components of the fuel injection system including the injector 17. Here, if the influence of the alcohol concentration on the air-fuel ratio feedback correction value faf can be completely ignored, the sum of the air-fuel ratio feedback correction value faf and the air-fuel ratio learning value KG is calculated from the rating of the fuel injection rate of the injector 17. It can be an index value of deviation. That is, if the fuel injection rate of the injector 17 becomes extremely large or small, the absolute value of the sum of the air-fuel ratio feedback correction value faf and the air-fuel ratio learning value KG becomes remarkably large. Therefore, in an internal combustion engine that always uses only fuel with a constant theoretical air-fuel ratio, it is possible to perform abnormality diagnosis of the fuel injection system using the sum of them as an index value efafkgd (= faf + KG).

  Further, even in an internal combustion engine using a mixed fuel with an undefined alcohol concentration, if the learning of the coefficient value kkinj of the alcohol concentration learning value is properly performed, abnormality diagnosis of the fuel injection system is performed using the index value efafkgd. Is possible. However, depending on the situation, it is conceivable that the coefficient value kkinj is not properly learned. In this case, the index value efafkgd reflects not only the deviation from the rating of the fuel injection rate of the injector 17 but also the amount corresponding to the learning error of the coefficient value kkinj, and the abnormality diagnosis of the fuel injection system is performed. Cannot be performed accurately.

  This will be described with a specific example. The horizontal axis of the graph in FIG. 2 indicates an index value efafkgd [%] obtained as the sum of the air-fuel ratio feedback correction value faf and the air-fuel ratio learning value KG. Here, the allowable range of the time lag required for fuel injection of the unit mass of the injector 17 is assumed to be “± 35%” of the rating. That is, in this case, if the change in the correction rate of the fuel injection time due to the change in the fuel injection rate of the injector 17 is within the range of “± 35%”, the fuel injection system is considered normal. become. Here, if the learning of the coefficient value kkinj of the alcohol concentration learning value is properly performed, only the influence of the change in the fuel injection rate of the injector 17 is reflected in the index value efafkgd. Therefore, at this time, if the index value efafkgd is within the range of “± 35%”, it can be determined that the fuel injection system is normal, and if the index value efafkgd is outside the range, the fuel injection system is abnormal.

  Consider a case where the coefficient value kkinj of the alcohol concentration learning value is erroneously learned. Considering the worst situation, there are the following two cases. That is, when the alcohol concentration of the actual fuel is the minimum “0%”, the coefficient value kkinj of the alcohol concentration learning value becomes a value “1.382” corresponding to the maximum alcohol concentration “85%” of the used fuel. When the actual alcohol concentration of the fuel is “85%” at maximum, the coefficient value kkinj of the alcohol concentration learning value is a value “1.0” corresponding to the minimum alcohol concentration “0%” of the fuel used. This is the case. In the former case, even if the fuel injection rate of the injector 17 is within the allowable range, the index value efafkgd can take a value in the range of “−35% to + 35%”. In the latter case, even if the fuel injection rate of the injector 17 is within the allowable range, the index value efafkgd may take a value in the range of “−9% to + 87%”. Therefore, the index value efafkgd may take a value in the range of “−35% to 87%” even if the change in the fuel injection rate of the injector 17 remains within the allowable range. Even in such a case, if the index value efafkgd deviates from the range of “−54 to 87%”, it can be determined that there is an abnormality in the fuel injection system. However, when the index value efafkgd is in the range of “35 to 87%”, it can be said that there is a possibility of abnormality, but it cannot always be determined that there is an abnormality.

  Here, there are the following two options for handling the normal and abnormal uncertain areas as described above. First, since there is a possibility of abnormality, it is considered that such an area is abnormal. In this case, if an abnormality occurs in the fuel injection system, this can be reliably detected. However, considering the frequency of occurrence of abnormalities in the fuel injection system and the frequency of mislearning of the coefficient value kkinj, even though there is actually no abnormality, a misdiagnosis that there is an abnormality frequently occurs. As a result, the reliability of the system is impaired.

  Therefore, as a second option, it can be considered that the above-mentioned area has no abnormality. In this case, it is determined that there is an abnormality only when there is an abnormality. However, there is no mislearning, and even when the index value efafkgd reaches “−54%” or “87%” only by a change in the fuel injection rate of the injector 17, it is not determined that there is an abnormality. . That is, in this case, the allowable range of the time lag required for fuel injection of the unit mass of the injector 17 is greatly expanded to “−54 to 87%” of the rating. As long as there is no abnormality, it is necessary to satisfy the regulations on emissions of internal combustion engines. Therefore, in this case, the emission guarantee range is greatly expanded. It should be noted that an extreme change in the fuel injection rate of the injector 17 has a great influence on the atomization of the fuel and the agitation property to the air, and can cause the combustion state of the internal combustion engine and thus the emission. Therefore, it is very difficult to establish the emission while allowing the deviation of the fuel injection rate over a wide range as described above.

  As described above, as long as the index value efafkgd is used, if an attempt is made to secure the reliability of the system, a large expansion of the emission guarantee range is inevitable. Therefore, in the present embodiment, an index value different from this is adopted to perform abnormality diagnosis of the fuel injection system, thereby suppressing the expansion of the emission guarantee range. The index value employed in the present embodiment has been proposed based on the following knowledge.

  In the tendency of deviation in the correction rate of the fuel injection time of the air-fuel ratio feedback control due to the erroneous learning of the coefficient value kkinj of the alcohol concentration learning value, a difference due to the coefficient value kkinj at that time is observed. For example, if the coefficient value kkinj is set to “1.382” which is the upper limit value of the possible range, there is a possibility that the value is overestimated, that is, that it is erroneously learned as a value larger than the original value. Even so, there is no possibility of the reverse. Therefore, the deviation of the correction rate of the fuel injection time of the air-fuel ratio feedback control due to the erroneous learning of the coefficient value kkinj at this time occurs only on the minus side (rich side). Also, if the coefficient value kkinj is set to “1.0” which is the lower limit value of the possible range, there is a possibility that the coefficient value kkinj is underestimated, that is, erroneously learned as a value smaller than the original value. Even so, there is no possibility of the reverse. Therefore, the deviation of the correction rate of the fuel injection time of the air-fuel ratio feedback control due to the erroneous learning of the coefficient value kkinj at this time occurs only on the plus side (lean side). When the coefficient value kkinj is set to an intermediate value between the upper limit value “1.382” and the lower limit value “1.0” of the possible range, the larger the value, the fuel that can be generated by the erroneous learning. The range of deviation of the correction rate of the injection time increases on the minus side (rich side) and decreases on the plus side (lean side).

  FIG. 3 shows the relationship between the coefficient value kkinj of the alcohol concentration learning value and the range of the deviation of the correction rate of the fuel injection time that may occur due to the erroneous learning. When the coefficient value kkinj of the alcohol concentration learning value is set to a value “1.382” corresponding to the alcohol concentration “85%” and the actual alcohol concentration is “85%”, that is, accurate learning is performed. If so, the fuel injection time correction factor (efafkgd = faf + KG) in the air-fuel ratio feedback control takes a value in the range of “± 35%” if there is no abnormality in the fuel injection system. When the coefficient value kkinj of the alcohol concentration learning value is “1.382” and the value is mislearned, the worst situation is when the actual alcohol concentration is “0%”. The fuel injection time correction factor (efafkgd = faf + KG) in the air-fuel ratio feedback control at this time may take a value in the range of “−54% to −3%” even if there is no abnormality in the fuel injection system. That is, the range of values that can be taken by the index value efafkgd when there is no abnormality in the fuel injection system when the coefficient value kkinj is “1.382” is a range of “−54% to + 35%”.

  On the other hand, when the coefficient value kkinj of the alcohol concentration learning value is set to a value “1.0” corresponding to the alcohol concentration “0%”, the actual alcohol concentration is “0%”, and accurate learning is performed. If so, the correction ratio of the fuel injection time in the air-fuel ratio feedback control takes a value in the range of “± 35%” if there is no abnormality in the fuel injection system. If the coefficient value kkinj of the alcohol concentration learning value is “1.0” and the value is mislearned, the worst situation is when the actual alcohol concentration is “85%”. The fuel injection time correction factor (efafkgd = faf + KG) in the air-fuel ratio feedback control at this time may take a value in the range of “−9% to + 87%” even if there is no abnormality in the fuel injection system. That is, the range of values that can be taken by the index value efafkgd when there is no abnormality in the fuel injection system when the coefficient value kkinj is “1.0” is a range of “−35% to + 87%”.

  When the coefficient value kkinj of the alcohol concentration learning value is between the minimum value “1.0” and the maximum value “1.382”, the index value efafkgd in the state where there is no abnormality in the fuel injection system is taken. The lower limit value and upper limit value of the range of values to be obtained shift to the minus side as the coefficient value kkinj increases. That is, the lower limit value and the upper limit value of the range of possible values of the index value efafkgd at each coefficient value kkinj are as shown by the curve L and the curve H in FIG. From such a relationship, the range of deviation of the correction rate of the fuel injection time that can be caused by the erroneous learning can be limited by the set value of the coefficient value kkinj at that time. In this case, the emission guarantee range can be narrowed by the area indicated by hatching in the figure as compared with the case where the second option is adopted.

  In order to realize normality / abnormality determination in this manner, the present embodiment adopts a new index value fakgalcd calculated by the following equation (2) instead of the index value efafkgd (= faf + KG). Therefore, the abnormality diagnosis of the fuel injection system is performed.

ちなみに、上式（２）は、係数値ｋｋｉｎｊ［倍］に代えて、アルコール濃度学習値（補正率）ＥＧ［％］（＝（ｋｋｉｎｊ−１）×１００）を用いると、下式（３）のように表すことができる。下式（３）に示されるように、指標値ｆａｆｋｇａｌｃｄは、空燃比学習値ＫＧとアルコール濃度学習値ＥＧとの積を用いて算出されている。

Incidentally, when the alcohol concentration learning value (correction rate) EG [%] (= (kkinj−1) × 100) is used instead of the coefficient value kkinj [times], the above equation (2) It can be expressed as As shown in the following equation (3), the index value fakgalcd is calculated using the product of the air-fuel ratio learned value KG and the alcohol concentration learned value EG.

なお、図３の設定例での診断での正常、異常の判定基準となる燃料噴射量補正率のマイナス側（リッチ側）、プラス側（リーン側）の閾値を示す曲線Ｌ、曲線Ｈはそれぞれ、上記指標値ｆａｆｋｇａｌｃｄが「−３５」、「＋８７」となるときの燃料噴射時間補正率の係数値ｋｋｉｎｊに応じた変化曲線となっている。したがって本実施の形態では、上記指標値ｆａｆｋｇａｌｃｄが「−３５」未満となるとき、或いは「＋８７」を上回るときに異常有りと判定して、燃料噴射系の異常診断を行うようにしている。

Note that curves L and H indicating threshold values on the minus side (rich side) and plus side (lean side) of the fuel injection amount correction factor, which are normal and abnormal judgment criteria in the diagnosis in the setting example of FIG. 3, respectively. , A change curve corresponding to the coefficient value kkinj of the fuel injection time correction factor when the index value fakgalcd is “−35” or “+87”. Therefore, in the present embodiment, when the index value fakgalcd is less than “−35” or exceeds “+87”, it is determined that there is an abnormality, and abnormality diagnosis of the fuel injection system is performed.

  FIG. 4 shows a flowchart of a “fuel injection system abnormality diagnosis routine” employed in this embodiment. The processing of this routine is periodically performed by the electronic control unit 28 during engine operation.

  When this routine is started, the electronic control unit 28 first determines in step S10 whether or not an abnormality diagnosis execution condition is satisfied. This implementation condition is that the operating condition of the internal combustion engine is stable, the rate of change of the air-fuel ratio feedback correction value faf is small to some extent, and learning of the air-fuel ratio learning value KG and the coefficient value kkinj of the alcohol concentration learning value is completed. And so on. And the electronic control unit 28 implements the process after step S20 only when the implementation condition is satisfied (S10: YES).

  In step S20, the electronic control unit 28 reads the air-fuel ratio feedback correction value faf, the air-fuel ratio learning value KG, and the coefficient value kkinj of the alcohol concentration learning value at that time. Then, in step S30, the electronic control unit 28 calculates the index value fafkgald based on the above equation (2). After the calculation, in step S40, the electronic control unit 28 determines whether or not the calculated index value fakgalcd is within the range of “−35 to 87”, and if it is within the range, the normal determination (S50) is performed. Otherwise, abnormality determination (S60) is performed, and the processing of this routine is terminated.

  In this embodiment, the coefficient value kkinj is the fuel concentration learning value obtained by learning the correction value of the fuel injection time corresponding to the difference in the fuel mixing ratio in the air-fuel ratio feedback control. The parameters correspond to the coefficient values of the fuel concentration learning value. In the present embodiment, the index value fafkgald calculated based on the above equation (2) is also a parameter corresponding to the index value of the correction degree of the fuel injection time in the air-fuel ratio feedback control. Further, by calculating the index value fakgalcd by reflecting the alcohol concentration learning value EG (coefficient value kkinj) based on the above formula (2), the normal / abnormal determination mode in diagnosis can be determined as the alcohol concentration learning value EG (coefficient value). The electronic control unit 28 that changes according to kkinj) corresponds to the determination mode changing means.

According to the present embodiment described above, the following effects can be obtained.
In the present embodiment, abnormality diagnosis of the fuel injection system is performed using the index value fafkgald calculated using the coefficient value kkinj of the alcohol concentration learning value based on the above equation (2). In this case, since the index value fafkgalcd used for the determination of normality / abnormality is operated according to the coefficient value kkinj of the alcohol concentration learning value, the normal / abnormal determination mode in the diagnosis is substantially the fuel concentration learning. It will be changed according to the value (coefficient value kkinj). By using such an index value fakgalcd, it is possible to minimize the expansion of the emission guarantee range while avoiding erroneous determination that there is an abnormality due to erroneous learning of the coefficient value kkinj. Therefore, according to the present embodiment, the abnormality diagnosis of the fuel injection system can be accurately performed while suppressing the expansion of the guaranteed emission range.

In addition, the said embodiment can also be changed and implemented as follows.
In the above embodiment, the index value fafkgald of the correction degree of the fuel injection time in the air-fuel ratio feedback control used for abnormality diagnosis of the fuel injection system is calculated based on the above equation (2). If the air-fuel ratio learning value KG is sufficiently learned and the air-fuel ratio feedback correction value faf is sufficiently small, the term of the air-fuel ratio feedback correction value faf in the above equation (2) is ignored. it can. Therefore, if the abnormality diagnosis is performed on the assumption that the air-fuel ratio learning value KG is sufficiently learned, the index value fafkgald is calculated using the above equation (2) with the term of the air-fuel ratio feedback correction value faf omitted. Calculation may be performed. In this case, the index value fakgalcd is a product of the air-fuel ratio learned value KG and the alcohol concentration learned value EG.

  In the above embodiment, the alcohol concentration learned value EG (its coefficient value kkinj) is reflected in the calculation of the index value fafkgalcd of the correction degree in the air-fuel ratio feedback control used for abnormality diagnosis of the fuel injection system. The normal / abnormal determination mode in the abnormality diagnosis is changed according to the alcohol concentration learning value EG. The same abnormality determination mode can be changed, for example, in the following mode. That is, an index value efafkgd (= faf + KG), which is a correction value of the fuel injection time TAU in the air-fuel ratio feedback control excluding the amount depending on the alcohol concentration learning value EG, is used as the index value of the correction degree. At the same time, the threshold value of the index value efafkgd, which is a normal / abnormal determination criterion in abnormality diagnosis, is changed according to the alcohol concentration learning value EG. Even in such a case, the normal / abnormal determination mode in the abnormality diagnosis can be changed according to the alcohol concentration learning value EG. In this case as well, the curves L and H in FIG. 3 are used as threshold values on the minus side (rich side) and plus side (lean side) of the fuel injection amount correction rate that is a criterion for determination of normality and abnormality in diagnosis. In addition, if the normality / abnormality is determined by comparing these with the index value efafkgd, the abnormality diagnosis of the fuel injection system can be performed in a manner substantially similar to the above embodiment.

  In the above embodiment, the case where the present invention is applied to an internal combustion engine that uses a mixed fuel of alcohol and gasoline has been described. However, the abnormality diagnosis apparatus and abnormality diagnosis method for a fuel injection system according to the present invention is also applied to an internal combustion engine using a mixed fuel other than a combination of alcohol and gasoline, which is composed of two types of fuels having different theoretical air-fuel ratio points. The present invention can be applied in a manner similar to or similar to the above-described form.

BRIEF DESCRIPTION OF THE DRAWINGS The schematic which shows typically the structure of the internal combustion engine applied about one Embodiment of this invention, and its control system. The graph which shows the relationship between the coefficient value kkinj of the alcohol concentration learning value in the same embodiment, and the correction | amendment rate of allowable fuel injection time. The graph which shows the relationship between the coefficient value kkinj of alcohol concentration learning value in the same embodiment, and the range of the deviation | shift of the correction rate of the fuel injection time which may arise by the mislearning. The flowchart which shows the process sequence of the abnormality diagnosis routine of the fuel-injection system applied to the embodiment.

Explanation of symbols

  DESCRIPTION OF SYMBOLS 10 ... Intake passage, 11 ... Combustion chamber, 12 ... Exhaust passage, 13 ... Air cleaner, 14 ... Air flow meter, 15 ... Throttle valve, 16 ... Intake port, 16a ... Intake manifold, 17 ... Injector, 18 ... Fuel pump, 19 ... Fuel tank, 20 ... delivery pipe, 21 ... intake valve, 22 ... spark plug, 23 ... exhaust valve, 24 ... exhaust port, 25 ... exhaust manifold, 26 ... oxygen concentration sensor, 27 ... three-way catalytic converter, 28 ... electronic control Unit (determination mode changing means), 29 ... accelerator sensor, 30 ... throttle sensor, 31 ... NE sensor, 32 ... remaining amount sensor.

Claims (8)

Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. In the fuel injection system abnormality diagnosis device for diagnosing based on the degree of correction of the fuel injection time of
Determination mode changing means for changing the normal / abnormal determination mode in the diagnosis according to the fuel concentration learned value by reflecting the fuel concentration learned value in the calculation of the correction degree index value in the diagnosis. Prepared,
In the internal combustion engine, a steady deviation of the fuel injection time in the feedback control is learned as an air-fuel ratio learning value,
The index value is calculated using a product of the air-fuel ratio learning value and the fuel concentration learning value.
An abnormality diagnosis apparatus for a fuel injection system. Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. In the fuel injection system abnormality diagnosis device for diagnosing based on the degree of correction of the fuel injection time of
The diagnosis is performed using a correction value of the fuel injection time in the feedback control excluding the amount depending on the fuel concentration learning value as an index value of the correction degree,
Normal before Symbol diagnosis, by changing according to the threshold of the fuel concentration learning value of the index value as the abnormality determination criteria, depending successfully in the diagnosis, the abnormality determination aspect to the fuel concentration learned value Determination mode changing means for changing
An abnormality diagnosis apparatus for a fuel injection system. Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. In the fuel injection system abnormality diagnosis device for diagnosing based on the degree of correction of the fuel injection time of
The steady deviation of the fuel injection time in the feedback control is learned as an air-fuel ratio learning value, the air-fuel ratio learning value is “KG [%]”, and the feedback correction value of the fuel injection time according to the deviation is “ faf [%] ", when the fuel concentration learning value is" EG [%] "
fafkgalcd = faf + EG × KG
An abnormality diagnosis device for a fuel injection system, wherein the diagnosis is performed by determining that the fuel injection system is abnormal when a value “fakgalcd” obtained by the following relational expression deviates from a specified normal range. Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. In the fuel injection system abnormality diagnosis device for diagnosing based on the degree of correction of the fuel injection time of
The steady deviation of the fuel injection time in the feedback control is learned as an air-fuel ratio learning value, the air-fuel ratio learning value is “KG [%]”, and the feedback correction value of the fuel injection time according to the deviation is “ faf [%] ”, the coefficient value of the fuel concentration learning value is“ kkinj [times] ”, the division value of the cylinder inflow air mass by the reference air-fuel ratio is“ K (KL) [g] ”, and the rated value of the fuel injection rate As the reciprocal of “α [sec / g]”,
TAU = α × K (KL) × (1 + FAF / 100 + KG / 100) × kkinj
When the value “TAU” obtained by the relational expression
fafkgalcd = faf + {kkinj × (1 + KG / 100) −1} × 100
An abnormality diagnosis device for a fuel injection system, wherein the diagnosis is performed by determining that the fuel injection system is abnormal when a value “fakgalcd” obtained by the following relational expression deviates from a specified normal range. Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. A method of diagnosing based on the correction degree of the fuel injection time of
By reflecting the fuel concentration learning value in the calculation of the index value of the correction degree in the diagnosis, the normal / abnormal determination mode in the diagnosis is changed according to the fuel concentration learning value,
In the internal combustion engine, a steady deviation of the fuel injection time in the feedback control is learned as an air-fuel ratio learning value,
Abnormality diagnosis method of fuel injection system you and calculates the index value using the product of the fuel concentration learned value and its air-fuel ratio learning value. Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. A method of diagnosing based on the correction degree of the fuel injection time of
Performing the diagnosis using the correction value of the fuel injection time in the feedback control excluding the amount depending on the fuel concentration learning value as an index value of the correction degree;
By changing the threshold value of the index value, which is a normal / abnormal judgment criterion in the diagnosis, according to the fuel concentration learned value, the normal / abnormal judgment mode in the diagnosis is changed according to the fuel concentration learned value. abnormality diagnosis method of fuel injection system you and changes. Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. A method of diagnosing based on the correction degree of the fuel injection time of
The steady deviation of the fuel injection time in the feedback control is learned as an air-fuel ratio learning value, the air-fuel ratio learning value is “KG [%]”, and the feedback correction value of the fuel injection time according to the deviation is “ faf [%] ", when the fuel concentration learning value is" EG [%] "
fafkgalcd = faf + EG × KG
An abnormality diagnosis method for a fuel injection system, wherein the diagnosis is performed by determining that the fuel injection system is abnormal when a value “fakgalcd” obtained by the following relational expression deviates from a specified normal range. Using a mixed fuel of two kinds of fuels having different theoretical air-fuel ratio points, and performing feedback control of the fuel injection time to reduce the deviation between the theoretical air-fuel ratio and the actual air-fuel ratio based on the detection result of the exhaust oxygen concentration, The feedback control is applied to an internal combustion engine that learns a fuel injection time correction value corresponding to a difference in the fuel mixture ratio as a fuel concentration learning value, and an abnormality in the fuel injection system of the internal combustion engine is detected by the feedback control. A method of diagnosing based on the correction degree of the fuel injection time of
The steady deviation of the fuel injection time in the feedback control is learned as an air-fuel ratio learning value, the air-fuel ratio learning value is “KG [%]”, and the feedback correction value of the fuel injection time according to the deviation is “ faf [%] ”, the coefficient value of the fuel concentration learning value is“ kkinj [times] ”, the division value of the cylinder inflow air mass by the reference air-fuel ratio is“ K (KL) [g] ”, and the rated value of the fuel injection rate As the reciprocal of “α [sec / g]”,
TAU = α × K (KL) × (1 + FAF / 100 + KG / 100) × kkinj
When the value “TAU” obtained by the relational expression is set as the fuel injection time,
fafkgalcd = faf + {kkinj × (1 + KG / 100) −1} × 100
An abnormality diagnosis method for a fuel injection system, wherein the diagnosis is performed by determining that the fuel injection system is abnormal when a value “fakgalcd” obtained by the following relational expression deviates from a specified normal range.

Images