JP5862511B2 - Vehicle learning data reuse determination device - Google Patents

Description

  The present invention relates to a vehicular learning data reuse determination device for reusing learned values.

  In general, an electronic control unit (ECU) is mounted on a vehicle, and the electronic control device performs various controls such as fuel injection control. When the electronic control device is abnormal or is likely to be abnormal, an inspection is performed at a store such as a dealer, and the electronic control device is replaced if necessary. The electronic control device stores a learning value based on data that is sequentially accumulated during traveling and when the vehicle stops. However, these learning values are not stored in a new electronic control unit.

  Therefore, if the learning value stored in the vehicle-mounted ECU can be reused, it is desirable that various data suitable for the vehicle can be used. Therefore, a technique is provided in which various types of information are stored in the electronic control unit, and the stored information is stored again in a new electronic control unit through communication means (see, for example, Patent Document 1).

  In addition, various types of vehicle-specific information are stored in the nonvolatile memory inside the ECU, and when the electronic control unit is replaced, the vehicle-specific information read from the electronic control unit before replacement is stored in the nonvolatile memory. There is also a technique in which a new history is written and the replacement history of the electronic control unit is written in a nonvolatile memory following the writing of the vehicle specific information (see, for example, Patent Document 2).

JP 07-103063 A JP 2000-185606 A

  However, when the above-described conventional technology is applied as it is to the stored information of the learning value of the vehicle, for example, the learning value that is mis-learned according to some other abnormality in the vehicle is reused. The subsequent data is also stored again in the new electronic control unit, which is not preferable.

  That is, in the technique according to the above-mentioned document, mis-learning data is reused without being sorted. For example, if a learning value related to engine control is used without being sorted, the emission characteristics of exhaust gas are reduced. There is a possibility that problems such as deterioration, deterioration of riding comfort, and deterioration of engine drivability may occur.

  An object of the present invention is to provide a vehicular learning data reuse determination device that enables only learning values necessary for vehicle control to be reused as much as possible.

Claim 1 or 2 SL placing of the invention, the learning values for the electronic control unit in the vehicle control before the replacement, and includes a storage means for storing diagnostic information of various abnormal as an abnormal another flag, the learning value in the storage means It is intended for technologies that determine whether or not can be reused. Further, in the vehicle learning data reuse determination device according to claim 1, an index code indicating an abnormality actually occurring in the vehicle by flag-on among the learning value and the diagnosis information through the scan tool from the electronic control device before replacement is provided. This is a new electronic control device that can be transferred. In addition, the vehicle learning data reuse determination device according to claim 2 is configured to be able to communicate with the electronic control device before the exchange through the in-vehicle network, and flags the learning value and the diagnosis information through the in-vehicle network. An index code indicating an abnormality that has actually occurred is configured in another electronic control unit that can be transferred.
According to the first aspect of the present invention, for example, a mechanic at a repair shop or the like transfers the flag value-on indicator code of the learning value and the diagnosis information from the electronic control device before replacement through the scan tool to the memory. The related storage means stores in advance a learning value index attached to the learning value and a diagnosis code that indicates an abnormality index that is an abnormality that makes the learning value unusable and may occur in the vehicle. and that although, determination means, associated storage means is stored in reference to the association between the learned value index and diag code, on whether an index code transferred instead was diagnostic information in the memory is included in the diagnostic code In response , the learning value stored in the storage means corresponding to the learning value index associated with the diagnosis code is transferred to a new electronic control unit to determine whether or not it can be reused .
According to the second aspect of the invention, the reading means transfers the learning value and the index code of the diagnostic information from the storage means of the electronic control device before the exchange to the memory through the in-vehicle network. The determination means refers to the association between the learning value index and the diagnosis code stored in the related storage means, and depending on whether or not the index code of the diagnosis information transferred to the memory is included in the diagnosis code, The learning value stored in the storage means corresponding to the learning value index associated with the diagnosis code is temporarily saved in another electronic control device and transferred to a new electronic control device that is replaced with the electronic control device before the replacement. To determine whether it can be reused.

  The determination means can derive a learning value index that is affected by the abnormality from the diagnosis information indicating the abnormality that has actually occurred, and can determine that the learning value corresponding to the derived learning value index cannot be reused. On the contrary, the determination means determines that the learning value can be reused when it is determined that there is no possibility that the influence of the abnormality is given to the learning value even if the diagnosis information is actually generated in the vehicle. Thereby, only the learning value necessary for vehicle control can be reused as much as possible.

BRIEF DESCRIPTION OF THE DRAWINGS The 1st Embodiment of this invention is shown, and the block diagram which shows roughly an example of the fuel-injection system in a motor vehicle Characteristic diagram showing the relationship between output voltage and intake air volume Flow chart schematically showing learning value calculation processing operation Schematic diagram showing reading and writing of learning values performed by the scan tool The figure which shows an example of the recording state of diagnostic information An example of a judgment table A flowchart schematically showing a process for determining whether or not a learning value can be reused FIG. 5 equivalent view showing the second embodiment of the present invention An example of a judgment table that stores combinations of diagnostic codes 7 equivalent diagram Example of learning value stored corresponding to each diagnosis code Injector flow ratio that changes according to machine difference FIG. 4 equivalent view showing another embodiment of the present invention

(First embodiment)
Hereinafter, a first embodiment of the present invention will be described with reference to FIGS. In the following description, for the intake air amount sensor that measures the intake air amount of an automobile engine, the learned value of the air-fuel ratio accompanying the aging of its output signal is transferred from the ECU (Electronic Control Unit) before replacement to a new ECU. A form in the case of reusing and replacing will be described.

  FIG. 1 shows an example of a fuel injection system in an automobile. The engine 1 is provided with an engine body 2, an injector 3 serving as a fuel injection valve, a fuel passage 4, an intake passage 5, an exhaust passage 6, a throttle valve 7, and the like. The fuel is pressurized by a high-pressure pump or the like and supplied to the injector 3 of each cylinder via the fuel flow path 4. The high-pressure fuel is injected from the injector 3 into the combustion chamber of the engine body 2 and is mixed with the intake air supplied via the intake passage 5 and the intake manipulator to form an air-fuel mixture.

  As shown in FIG. 1, the engine 1 includes various sensors such as an intake air temperature sensor 8, an intake air pressure sensor 9, and an intake air amount sensor 10 in an intake passage 5 on the intake side of the engine body 2. The intake air temperature sensor 8, the intake air pressure sensor 9, and the intake air amount sensor 10 detect the temperature, pressure, and flow rate of the intake air flowing through the intake passage 5, and the detected intake air temperature, pressure, and flow rate are respectively used as electrical signals. Output to an ECU (Electronic Control Unit) 11. Note that the intake air temperature sensor 8 may be provided in the intake air amount sensor 10.

  The engine body 2 is provided with a cylinder block, a cylinder, and a piston (none of which are shown), and the mixture is exploded and expanded at the timing when the mixture is injected into the combustion chamber and the piston moves up. When the piston moves down, the crankshaft rotates.

  As shown in FIG. 1, sensor signals are given to the ECU 11 from various sensors such as a rotation speed sensor 12, a water temperature sensor 13, and an accelerator opening sensor 14. The rotation speed sensor 12 detects the rotation angle or / and rotation speed of the crankshaft and outputs it to the ECU 11 as an electrical signal. The water temperature sensor 13 is provided in the cylinder block of the engine 1, detects the cooling water temperature of the engine 1, and outputs it to the ECU 11 as an electrical signal. The accelerator opening sensor 14 detects the opening of an accelerator pedal (not shown) and outputs it to the ECU 11 as an electrical signal.

  Exhaust gas discharged from the exhaust valve of the engine 1 is discharged to the exhaust passage 6 through the exhaust manifold. An air-fuel ratio sensor 15 is provided in the exhaust passage 6. The air-fuel ratio sensor 15 detects the air-fuel ratio of the exhaust gas and outputs it to the ECU 11 as an electrical signal.

  The ECU 11 is mainly composed of a CPU 16, and in addition, a ROM 17 for storing tables of various programs, flags, etc., a RAM 18 for temporarily storing flag values and calculation data corresponding to the tables stored in the ROM 17, an SRAM 19, A memory (hereinafter referred to as a memory as necessary) such as an EEPROM 20 serving as a nonvolatile memory is provided. The SRAM 19 is used as a backup RAM for holding stored data using a battery as a power source even when the engine is stopped.

  Various programs are stored in the ROM 17, and the CPU 16 of the ECU 11 performs various controls of the engine 1 by executing the stored programs in the ROM 17. When the CPU 16 executes the control program in the ROM 17, it learns about various engine controls, and stores the learned values in a memory for use in later engine control. In general, when the automobile is a new car, the ECU 11 holds the learning value as an initial value. When the engine 1 is controlled, the learning values are sequentially stored or rewritten.

  Hereinafter, the features of the present embodiment will be exemplified and described. Specifically, the case where the air-fuel ratio learning value used when calculating the injection time for injecting fuel from the injector 3 is transferred to a new ECU will be described as an example. This air-fuel ratio learning value is stored or rewritten in the aforementioned SRAM 19 and / or EEPROM 20.

  As shown in the output voltage-intake air amount characteristic of FIG. 2, the output voltage of the intake air amount sensor 10 uniquely changes according to the intake air amount. Therefore, the ECU 11 can obtain the actual intake air amount by detecting the output voltage value. However, even if the output voltage detected by the intake air amount sensor 10 is equivalent, the detected intake air amount changes when the intake air amount measurement system deteriorates over time. Referring to the design median characteristic C in FIG. 2 and the aged deterioration characteristic D, even if the output voltage is equal, the design median characteristic C is calculated by the value of the intake air amount A. On the other hand, in the case of the characteristic D at the time of deterioration, the intake air amount B is larger than the intake air amount A. That is, even if the output voltage of the intake air amount sensor 10 is equal, the actual intake air amount corresponds to a large value when the sensor deteriorates.

  As shown in the injection time calculation flowchart in FIG. 3, the CPU 16 of the ECU 11 uses the intake air amount data detected by the intake air amount sensor 10, the water temperature data detected by the water temperature sensor 13, and the rotation speed sensor 12. The basic injection time is calculated using the detected rotational speed data (S1). This basic injection time indicates the injection time before correction.

  Next, the CPU 16 of the ECU 11 calculates an air-fuel ratio correction value (S2). A value obtained by dividing the air-fuel ratio detected by the air-fuel ratio sensor 15 by the target air-fuel ratio (for example, 14.6) is calculated as the air-fuel ratio correction value. If the air-fuel ratio detected by the air-fuel ratio sensor 15 is detected to be lower than the target air-fuel ratio, the air-fuel ratio correction value becomes a value lower than 1, and if vice versa, the air-fuel ratio correction value becomes a value higher than 1.

  Then, the ECU 11 refers to the air-fuel ratio correction value and determines whether the lean state where the air-fuel mixture is thinner than the target air-fuel ratio or the rich state where the air-fuel mixture is rich (S3). Then, the ECU 11 determines whether the lean state or the rich state continues for a predetermined time or more. If this condition is not satisfied (S3: NO), the basic injection time (air-fuel ratio learning value + air-fuel ratio) is satisfied. The time multiplied by (correction value) is calculated as the injection time, and used as the command value for the injection time of the injector 3 (NO in S3 → S5).

  For example, when the engine is started, the internal combustion engine is not warmed up and the air-fuel ratio sensor 15 is not activated. Therefore, even if the air-fuel ratio correction value calculated from the air-fuel ratio sensor 15 is used, an ideal target air-fuel ratio ( For example, it is difficult to control to 14.6). Therefore, the actual injection time is set in step S5 so as to correct an error occurring at the time of start-up that cannot be corrected only by the air-fuel ratio correction value using the parameter of the air-fuel ratio learning value, and to approach the target air-fuel ratio also at the time of engine start. Has an influence.

  In the memory of the new ECU 11, the initial value of the air-fuel ratio learning value is set to a predetermined value and stored. Note that the air-fuel ratio learning value changes according to various parameters surrounding the engine 1 as well as the influence of the output value of the air-fuel ratio sensor 15. As time passes after the initial use of the engine 1, the injection control of the engine 1 also becomes substantially constant and the air-fuel ratio learning value tends to settle to a constant value.

  As described above, if the air-fuel ratio learning value is set in the ECU 11, the injection time is calculated by multiplying the basic injection time by (air-fuel ratio correction value + air-fuel ratio learning value). Since the air-fuel ratio correction value is calculated using the air-fuel ratio detected by the air-fuel ratio sensor 15, the injection time can be feedback controlled according to the actually detected air-fuel ratio.

  However, as described above, when the intake air amount sensor 10 deteriorates over time, the output voltage of the intake air amount sensor 10 deviates from the design median value, and the intake air amount is detected differently from the actual value. For this reason, the change over time of the output voltage of the intake air amount sensor 10 affects the basic injection time calculated in step S1. Actually, an offset corresponding to the secular change is added to the air-fuel ratio correction value. Then, at the determination timing of step S3, the lean state or the rich state continues for a predetermined time or more due to the influence of the offset (S3: YES). The lean state continues because the recognized flow rate of the ECU 11 is recognized to be less than the actual flow rate due to the secular change.

  Therefore, in the learning control of the present embodiment, when the determination result in step S3 is YES, an air-fuel ratio correction value is obtained by adding or subtracting (adding or subtracting) a predetermined offset value α to the air-fuel ratio learning value. Then, the air-fuel ratio learning value and the air-fuel ratio correction value are corrected by subtracting (subtracting or adding) the offset value α in the reverse direction (S4).

  The offset value α corresponds to aged deterioration. When it is determined in step S3 that the rich state continues, the offset value α is set to a positive value, and when it is determined that the lean state continues, the offset value α is set to a negative value. Set and add / subtract / subtract.

  In the present embodiment, the air-fuel ratio learning value is a learning value that takes into account an offset value corresponding to the secular change of the output voltage of the intake air amount sensor 10. By changing this offset value, it is possible to reflect the influence of the signal of the intake air sensor 10 over time.

  Now, let us assume a case where it becomes necessary to replace the ECU 11 that stores such an air-fuel ratio learned value in a memory due to some influence. For example, when the ECU 11 is replaced with a new ECU 30, an air-fuel ratio learning value is stored as an initial value in a memory mounted on the new ECU 30, and the air-fuel ratio learning value that has been mounted on the vehicle and has changed over the years has been stored. It is not reflected in the new air-fuel ratio learning value.

  However, when the air-fuel ratio learning value data is transferred from the SRAM 19 or the EEPROM 20 of the ECU 11 mounted on the vehicle to the memory of the new ECU 30, the new ECU 30 uses the air-fuel ratio learning value to command the injection time. The value can be calculated, and the influence of the error accompanying the secular change of the intake air amount by the intake system device such as the intake air amount sensor 10 can be reflected.

  Therefore, as shown in FIG. 4, for example, a scan tool 50 is prepared in a repair shop installed in an automobile dealer or the like. The scan tool 50 is operated by a mechanic or the like who performs car maintenance, reads all the learning values (in this embodiment, the air-fuel ratio learning value) from the SRAM 19 or the EEPROM 20 of the ECU 11, and reads the read learning values. This is a tool that can be directly written into a new ECU 30.

  The scan tool 50 also transfers the diagnosis information 33 together with all the learning values described above from the memory in the ECU 11 to the memory in the ECU 30. This diagnostic information 33 is a variety of in-vehicle sensors such as an intake air temperature sensor 8, an intake air pressure sensor 9, an intake air amount sensor 10, a rotation speed sensor 12, a water temperature sensor 13, an accelerator opening sensor 14, an air-fuel ratio sensor 15, and an injector 3. This is information indicating various abnormalities such as disconnection of the communication line extending to the actuator / power supply short circuit, etc., and recorded as diagnostic information 33 in the SRAM 19 or / and EEPROM 20 of the ECU 11 when an abnormality related to these sensors and actuators occurs. Has been.

  The diagnostic information 33 shown in FIG. 5 indicates that the flag is recorded for each block divided into intake air amount relationship, rotation speed relationship, water temperature relationship, intake air temperature relationship, injector relationship, and air-fuel ratio sensor relationship. . For example, the diagnostic information relating to the intake air amount stores various abnormalities such as a disconnection or short circuit of the communication line of the intake air sensor 10 and a power supply short circuit of the communication line as a flag for each abnormality.

  Further, the diagnosis information 33 relating to the rotational speed holds various abnormalities such as a disconnection or short circuit of the communication line of the rotational speed sensor 12 as a flag. Further, the diagnosis information 33 relating to the water temperature holds various abnormalities such as a disconnection or short circuit of the communication line of the water temperature sensor 13 and a power supply short circuit of the communication line as a flag for each abnormality.

  Further, the diagnostic information 33 relating to the intake air temperature holds various abnormalities such as disconnection and short of the communication line of the intake air temperature sensor 8 as a flag for each abnormality. Further, the diagnostic information 33 related to the injector 3 is held with an abnormality in the energization time for driving the injector 3 as a flag. The diagnosis information 33 related to the air-fuel ratio sensor 15 holds various abnormalities such as disconnection of the communication line of the air-fuel ratio sensor 15 and output voltage abnormality as a flag for each abnormality. The scan tool 50 can transfer a diag code (corresponding to an index code) flagged on in the diag information 33 from the ECU 11 to the SRAM 31 or the EEPROM 32 of the ECU 30.

The ECU 30 has the same configuration as the ECU 11. That is, the ECU 30 includes a RAM 34, a ROM 35, and a CPU 36 along with the SRAM 31 and the EEPROM 32.
A determination table 37 shown in FIG. 6 is recorded in the memory (for example, ROM 35) of the new ECU 30. This determination table 37 reuses the learning value index (learning value L1, learning value L2, learning value L3,...) Of each assumed learning value (in this embodiment, the air-fuel ratio learning value is an example) and the learning value. This indicates a correspondence with a diagnosis code (for example, P0102, P0103, etc.) of the diagnosis information to be disabled, and one or a plurality of diagnosis codes are recorded in association with each of the learning value indices L1 to L3.

  For example, the air-fuel ratio learning value determination table 37 includes a disconnection of the communication line extending to the intake air amount sensor 10, a power supply short circuit of the communication line extending to the intake air amount sensor 10, and a ground short circuit of the communication line extending to the intake air amount sensor 10. Abnormalities relating to various intake air amounts are recorded as diagnosis codes (P0102, P0103, etc.).

  The ROM 35 of the ECU 30 stores a learning value reusability determination program. The CPU 36 of the ECU 30 uses the determination program stored in advance in the ROM 35 to determine whether the learning value transferred to the SRAM 31 or the EEPROM 32 can be reused. to decide.

  FIG. 7 shows an outline of the reusability determination process performed by the CPU 36 of the ECU 30. The process of this flowchart is repeated by the number N of learning values (total number of learning values) in the determination table 37 (T1, T2). As shown in FIG. 7, it is determined whether or not the index code of the diag information that has actually occurred matches one of the diag codes recorded in the determination table 37 (T3). When there is an index code for diagnostic information, it is rewritten to an initial value preset in the ECU 30 (T5).

  For example, when “P0100” and “P0123” are included as index codes of diagnostic information indicating an abnormality occurring in the vehicle, the CPU 36 of the ECU 30 refers to the learning value determination table 37 shown in FIG. Since it can be confirmed that “P0100” is included in “Learning Value L2” and “P0100” and “P0123” are included in “Learning Value L3”, both “Learning Value L2” and “Learning Value L3” are erroneously learned. In step T5, “learned value L2” and “learned value L3” are rewritten to initial values.

  As a result, if any diagnosis that makes the learning value unusable has occurred, the learning value data stored in the ECU 11 before replacement is discarded, and the ECU 30 sets a predetermined initial value. Like to do.

  On the contrary, even if the index code of the diagnosis information is stored as flag-on in the ECU 11 before replacement, if there is no possibility that the influence of the abnormality is given to the learning value data, it is associated in the determination table 37. Therefore, the CPU 36 of the ECU 30 determines that the learning value data stored in the ECU 11 before replacement can be reused. That is, the learning value is rewritten in the ECU 30 and is not returned to the initial value.

  For example, in the above description, since “P0100” and “P0123” are not included in the determination table 37 of “learning value L1”, it is determined that the learning value is not related to the diagnosis and is rewritten by the scan tool 50. Keep the learned value.

  The air-fuel ratio learning value will be described as a specific example. When an abnormality such as disconnection of a communication line extending to the intake air amount sensor 10 or a power supply / ground short-circuit occurs, a diagnosis related to the intake air amount is generated. In this case, the reliability of the value of the output signal of the intake air amount sensor 10 becomes poor. Accordingly, it is predicted that the air-fuel ratio learning value obtained using the signal value is not a correct value and the reliability is poor. Become. Therefore, it is desirable that the air-fuel ratio learned value stored in the ECU 11 before replacement has a low utility value and the learned value is discarded. At this time, the air-fuel ratio learning value data stored in the ECU 11 before replacement is discarded, and an initial value preset in the new ECU 30 is used.

  Conversely, even if a diagnosis that is not related to the air-fuel ratio learning value occurs, the influence of the abnormality does not affect the stored value of the air-fuel ratio learning value, so the scan tool 50 is stored in the ECU 11 before replacement. The air-fuel ratio learned value data is transferred to the ECU 30 and used as it is. Thereby, only the learning value necessary for vehicle control can be reused as much as possible. In other words, even after replacement with a new ECU 30, the possibility of deterioration of emission characteristics of exhaust gas, deterioration of riding comfort, deterioration of engine drivability, and the like can be suppressed as much as possible.

  According to the present embodiment, the CPU 36 of the new ECU 30 has at least one of the index code of the diagnostic information actually generated in the vehicle on which the ECU 11 before replacement is mounted and the diagnostic code stored in the determination table 37 of the ECU 30. It is determined that the learning value corresponding to the diag code cannot be reused on condition that one or more match. As a result, it is possible to appropriately determine the learning value to be reused, and the learned value that has been learned in error can be discarded without being used in the new ECU 30. Further, the reusable learning value can be reused in the new ECU 30.

(Second Embodiment)
FIGS. 8 to 10 show a second embodiment of the present invention. The difference from the previous embodiment is that the combination of at least two of the index codes of the diagnosis information matches the diagnosis code. The learning value is judged not to be reusable. Parts that are the same as or similar to those in the previous embodiment are given the same or similar reference numerals, and descriptions thereof are omitted. Hereinafter, different parts will be described.

  As shown in FIG. 8, index code flags of various diagnostic information are stored for each block (intake air amount relationship, rotation speed relationship, water temperature relationship, intake air temperature relationship, injector relationship, air-fuel ratio sensor relationship). The diagnosis information includes, for example, a diagnosis flag that is generated when the detected water temperature is abnormally increased / abnormally lowered (for example, the standard water temperature ± 10%) by the water temperature sensor 13, and the abnormal intake air temperature detected / abnormally detected by the intake air temperature sensor 8. Also included is a diagnostic flag that is generated when the vehicle descends (eg, standard temperature ± 10%). Such a diagnosis may flag on.

  However, such an abnormal rise / fall of sensing is minor as an abnormal state. Therefore, when multiple combinations of diagnostics are generated, the learning value cannot be reused. It may be determined that the learning value can be reused. Then, even when a serious abnormality occurs due to a combination of minor abnormalities, the learning value related to the abnormality can be discarded, and conversely, if there is only a minor abnormality, the learning value related to the abnormality can be reused.

  Specifically, as shown in FIG. 9, a determination table 38 that replaces the determination table 37 of the above-described embodiment may store combinations of diag codes (for example, P0117 & P0112, P0118 & P0113).

  Then, as shown in the flowchart of FIG. 10 instead of FIG. 7 of the above-described embodiment, the CPU 36 of the ECU 30 refers to the determination table 37 (T1, T2, T4, T5), and the combination of the index codes of the diagnostic information that has actually occurred It is good to determine whether or not it matches the diagnosis code stored in the determination table 37 (T3a).

  For example, as shown in FIG. 8, the diagnosis code for an abnormal rise in water temperature (for example, standard temperature + 10% or more) is P0117, and the diagnosis code for an abnormal increase in intake temperature (for example, standard temperature + 10% or more) is set to P0112. It is assumed that both these flags are turned on. These combinations (P0117 & P0112) are stored in the determination table 37 shown in FIG. Therefore, the CPU 36 of the ECU 30 determines YES in step T3a, determines that the learning value data stored in the ECU 11 before replacement is not reusable, discards it, and rewrites the initial value preset in the new ECU 30. (T5).

  Conversely, even if a diagnosis has occurred, if there is no combination of the diagnosis codes in the determination table 37, it is determined that the corresponding learning value can be reused, and the learning value rewritten by the scan tool 50 Is kept as it is. Then, the highly reliable learning value stored in the ECU 11 before replacement can be reused.

  According to the present embodiment, the learning value is determined not to be reusable on condition that at least two combinations of the index codes indicated by the diagnosis information match the diagnosis code in the determination table 37. Even if a serious abnormality is caused by a combination of minor abnormalities, the learning value related to the abnormality can be discarded. Conversely, for example, even if only one minor diagnosis matches, it is determined that the learning value can be reused. Therefore, if there is only a minor abnormality, the learning value related to the abnormality can be reused. This makes it possible to more appropriately determine whether or not the learning value can be reused.

(Third embodiment)
FIG. 11 shows a third embodiment of the present invention. FIG. 11 shows a determination table 39 in place of the determination table 37 or 38. The determination table 39 stores a plurality of learning values L1, L3, L4, L5, L6,... Corresponding to each of the diagnosis codes (P0101). In addition, the determination table 39 stores a plurality of learning values L2, L5, L6, L8, L15,... Corresponding to the diagnosis code (P0211), and stores a plurality of learning values corresponding to the diagnosis code (P1001). L3, L7, L10, L13, L15,... Are stored in association with each other. Then, the CPU 36 of the new ECU 30 can immediately determine that all the learning values related to one diagnostic code cannot be reused.

  For example, if it is determined that the learning value L1 is not reusable, it is determined that the learning value L3 associated with the learning value L1 is not reusable in conjunction with the learning value L1. Anyway. That is, whether or not the learning value can be reused may be determined from the correlation between the learning values. Then, the determination can be made reliably and quickly.

(Other embodiments)
The present invention is not limited to the above-described embodiment, and for example, the following modifications or expansions are possible.
As an example of the learning value described in the above-described embodiment, an embodiment in which the learning value is applied to the air-fuel ratio learning value is shown. You may apply to other learning values, such as a learning value of an electric current value.

  For example, if a deposit adheres to the electronic throttle valve 7, the deposit is less than the intake air amount corresponding to the actual opening / closing state of the electronic throttle valve 7. Since the recognition flow rate by the ECU 11 is set according to the open / close control state of the electronic throttle valve 7, the recognition flow rate of the ECU 11 is detected more than the actual flow rate. For this reason, the intake air amount becomes smaller than the appropriate flow rate, and the engine speed decreases. Therefore, the flow rate learning value is used as the correction value, and the flow rate obtained by adding the flow rate learning value to the actual flow rate is recognized as the ECU 11 recognition flow rate.

  When an abnormality related to the flow rate learning value such as disconnection or short circuit of various communication lines is detected in the ECU 11 and stored in the SRAM 19 or the EEPROM 20, the flow rate learning value related to ISC (Idle Speed Control) may be erroneously learned. There is. Therefore, it is preferable not to reuse the learning value in the new ECU 30 after replacement.

  Further, the relative injection amounts of the large number of injectors 3 show a distribution as shown in FIG. The basic injection time obtained in the above embodiment is calculated from the design median value of the injector 3 (median value of distribution).

  The machine difference between the individual injectors 3 is detected as an air-fuel ratio using an air-fuel ratio sensor 15. Since the engine control system is feedback controlled using the detected air-fuel ratio as shown in the above-described embodiment, the machine difference of the injector 3 is also reflected in the air-fuel ratio correction value. If the actual injection amount of the injector 3 deviates significantly from the design median value, the air-fuel ratio correction value has an offset of the machine difference of the injector 3, and this state is shown in step S3 of FIG. If it continues for a predetermined time, the offset value α is added to or subtracted from the air-fuel ratio learning value.

  When the ECU 11 is replaced without replacing the injector 3, the air-fuel ratio learning value affected by the machine difference of the injector 3 is returned to the initial value as in the above-described embodiment. For this reason, when the ECU 11 is replaced with a new ECU 30, it is preferable to rewrite the air-fuel ratio learning value to the memory of the new ECU 30, but at this time as well, if any abnormal diagnosis occurs as described above, the air-fuel ratio The reliability of the learning value is lowered. Therefore, as shown in the above-described embodiment, it is desirable to determine whether reuse is possible using the determination table 37 or the like. In addition, the present invention can be similarly applied to learning values relating to the rotational speed, water temperature, and air-fuel ratio sensor.

  In the description of the above-described embodiment, the scan tool 50 reads the learning value data from the ECU 11 before replacement, writes the learning value to the new ECU 30, and the CPU 36 of the new ECU 30 determines whether or not the learning value can be reused. Although shown, this determination means may be incorporated in an external tool such as the scan tool 50.

  That is, since the scan tool 50 also reads the diagnosis information actually generated in the vehicle, if the determination table 37 or the like is stored in an internal memory (not shown) of the scan tool 50, the learning value is read. This is because it is sometimes possible to determine whether or not the reuse is possible.

  Further, in such a case, the embodiment has been described in which the learning value of the SRAM 19 or the EEPROM 20 of the ECU 11 is transferred to the SRAM 31 or / and the EEPROM 32 of the new ECU 30 to determine whether or not it can be reused. After determining the availability, only the reusable learning value may be transferred to the new ECU 30.

  Further, as shown in FIG. 13, in the vehicle, the ECU 11 (engine ECU) is connected to other ECUs (for example, brake ECU, transmission ECU) 52 via an in-vehicle network 51, and a plurality of ECUs connected to the in-vehicle network. Communication between them is possible.

The other ECU 52 connected to the ECU 11 via the in-vehicle network 51 has its CPU 36 (corresponding to the reading means) read the learning value stored in the ECU 11 through communication processing through the in-vehicle network 51 and is actually generated in the vehicle. The diagnosis information may be read from the ECU 11 to determine whether or not the read learning value can be reused. That is, the learning value may be temporarily saved in another ECU 52, and after the ECU 11 is replaced with a new ECU 30, the learning value stored in the other ECU 52 may be rewritten to the new ECU 30.

  That is, a device (vehicle learning data reuse determination device) including a determination unit (CPU 36) for determining whether or not a learning value can be reused and a related storage unit (SRAM 31 or / and EEPROM 32) is included in the new ECU 30 described in the above embodiment. It is not limited, and may be the scan tool 50, another ECU 52 in the vehicle, or the like. Further, the related storage means (SRAM 31 and / or EEPROM 32) and the determination means (CPU 36) may be provided in separate devices.

In the drawing, 10 is an intake air amount sensor, 11 is an ECU (electronic control device before replacement), 15 is an air-fuel ratio sensor, 19 is SRAM (storage means of the electronic control device before replacement), and 20 is EEPROM (electronic before replacement). Control unit storage means), 30 ECU (new electronic control unit), 31 SRAM (related storage means), 32 EEPROM (related storage means), 33 diagnostic information, 35 ROM, 36 CPU (determination) Means , reading means ), 37 to 39 are judgment tables, 50 is a scan tool, 51 is an in-vehicle network, and 52 is another ECU (other electronic control unit).

Claims (9)

The electronic control unit (11) before replacement includes a learning value relating to vehicle control, and storage means (19 or / and 20) for storing diagnostic information using various abnormalities as flags for abnormality , and uses the learning values. And
New electronic control in which the learning code and the indicator code indicating the abnormality actually occurring in the vehicle can be transferred from the electronic control device (11) before the replacement through the scan tool (50). A vehicle learning data reuse determination device configured in the device (30),
Through the scan tool from the electronic control device before the exchange, the learning value, and a memory in which the flag-on indicator code of the diagnostic information is transferred,
Associated with learning value indicator (L1 to L3) to be subjected to the studies習値, a diagnosis code indicating an abnormality of the indices a likelihood in the vehicle a abnormalities impossible reusing the learning value, the and associated storage means (31 or / and 32) you previously stored Te,
Said associated memory means (31 or / and 32) is stored in reference to the association between the learned value index and diag code, whether or not the index code transferred instead was da Iagu information in the memory is included in the diagnostic code Depending on whether or not, the learning value stored in the storage means (19 or / and 20) of the electronic control device before exchange corresponding to the learning value index associated with the diagnostic code is transferred to the new electronic control device. transferred instead by a determining means for determining whether the reusable (36), a vehicle learning data reuse determination apparatus comprising: a. The electronic control unit (11) before replacement includes a learning value relating to vehicle control , and storage means (19 or / and 20) for storing diagnostic information using various abnormalities as flags for abnormality, and uses the learning values. And
An index indicating an abnormality that is actually generated in the vehicle by flag-on among the learning value and the diagnosis information through the in-vehicle network, which is configured to be able to communicate with the electronic control device (11) before the replacement. A vehicle learning data reuse determination device configured in another electronic control device (52) to which a code can be transferred,
Memory,
Reading means (36) for transferring the flag value on the learning value and the diagnosis information from the electronic control device before the exchange through the in-vehicle network, to the memory;
The learning value index (L1 to L3) attached to the learning value is associated with a diagnosis code indicating an abnormality index that is an abnormality that makes the learning value non-reusable and may occur in the vehicle in advance. you storage associated storage means (31 or / and 32),
Said associated memory means (31 or / and 32) is stored in reference to the association between the learned value index and diag code, whether or not the index code transferred instead was da Iagu information in the memory is included in the diagnostic code Depending on whether or not, the learning value stored in the storage means (19 or / and 20) of the electronic control device before exchange corresponding to the learning value index associated with the diagnostic code is used as the other electronic control device. and determining means for determining whether the reusable transferred to the exchange before the electronic control unit new electronic control device to be exchanged (30) with temporarily saved in the (52) (36), the A vehicle learning data reuse determination device comprising: a vehicle learning data reuse determination device; The determination means (36) uses the learning value index referred to as a condition on the condition that at least one of the index code of the diagnostic information indicating the abnormality actually occurring in the vehicle matches the diagnostic code of the related storage means. 3. The learning data for a vehicle according to claim 1 or 2, wherein it is judged that the learning value of the storage means (19 or / and 20) of the corresponding electronic control unit (11) before the replacement cannot be reused. Usage judgment device. The determination means (36) is configured to change the electronic control device (11) before replacement when the index code of the diagnosis information indicating an abnormality that has occurred in the vehicle matches only one of the diagnosis codes of the related storage means. The learning value of the storage means (19 or / and 20) is determined to be reusable,
On the condition that the combination of at least two index codes indicated by the diagnostic information matches the diagnostic code of the related storage means (31 or / and 32), the pre-exchange data corresponding to the referenced learning value index 3. The vehicle learning data reuse determination device according to claim 1, wherein it is determined that the learning value of the storage means (19 or / and 20) of the electronic control device (11) cannot be reused. The vehicle learning according to any one of claims 1 to 4, wherein the relation storage means (31 or / and 32) stores a plurality of diagnosis codes in association with each learning value index. Data reuse judgment device. The vehicle learning according to any one of claims 1 to 4, wherein the association storage means (31 or / and 32) stores a plurality of learning value indexes in association with each diagnosis code. Data reuse judgment device. The vehicle learning according to any one of claims 1 to 6, wherein the determination means (36) determines whether or not the learning value can be reused from a correlation between a plurality of learning values. Data reuse judgment device. When the determination means (36) determines that the learning value relating to vehicle control cannot be reused, the learning value relating to vehicle control is not discarded and used, and the initial value is held in the new electronic control unit (30). The vehicle learning data reuse determination device according to any one of claims 1 to 7, wherein: The electronic control device (11) before the exchange is
A function of calculating the basic injection time according to the intake air amount detected by the intake air amount sensor (10) and a value obtained by dividing the detected air-fuel ratio detected by the air-fuel ratio sensor (15) by the target air-fuel ratio. A function to calculate as a value, a function to determine whether the lean state or the rich state continues for a predetermined time or longer according to the air-fuel ratio correction value, and a predetermined value to the air-fuel ratio learning value if it is determined to continue for a predetermined time or longer. A value obtained by adding or subtracting a value and subtracting or adding the predetermined value in the reverse direction to the air-fuel ratio correction value and multiplying the basic injection time by (air-fuel ratio correction value + air-fuel ratio learning value) is calculated as an injection time. A storage means (19 or / and 20) of the electronic control unit (11) before replacement stores the air-fuel ratio learning value as the learning value.
The determination means (36) is configured to change the electronic control device (pre-replacement) according to an index code of diagnostic information generated due to a disconnection of a communication line connected to the intake air amount sensor (10) or a power supply / ground short. The vehicle learning according to any one of claims 1 to 8, wherein it is determined that the reuse of the air-fuel ratio learning value stored in the storage means (19 or / and 20) of 11) is impossible. Data reuse judgment device.

Images