JP2014051893A - Injector driving device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an injector driving device which detects continuous ON-mode of the injector precisely and quickly.SOLUTION: An injector driving device includes: injector control means which opens or closes a valve of an injector by controlling electric conduction to the injector and, thereby, controls fuel ejection of from the injector; electric conduction state abnormality determining means (S110) which determines abnormality of the state of electric conduction to the injector; and abnormal ejection state determining means (S140) which determines whether such an abnormal ejection state that fuel is abnormally ejected from the injector exists or not based on a change of fuel pressure. Therein, when it is determined that the abnormality of the electric conduction state occurs and it is determined that the abnormal ejection state exists, it is determined that such a continuous valve opening abnormal state (continuous ON-mode) that the injector is retained in a valve-opening state continues (S170).

Description

  The present invention relates to an injector driving device that drives an injector by controlling energization to the injector.

  2. Description of the Related Art Conventionally, there is known a common rail fuel injection system in which fuel pressurized by a fuel supply pump is accumulated in a pressure accumulating chamber (common rail) and injected into each cylinder (cylinder) of an engine from a fuel injection valve (injector). . Further, as an injector used in such a common rail fuel injection system, a so-called piezo injector that uses a piezo element that expands or contracts by charging and discharging and opens and closes by expansion and contraction of the piezo element is known.

  In such a fuel injection system, there is a possibility that an abnormality (hereinafter, also referred to as “continuous on mode”) occurs in which the injector is maintained in the valve open state regardless of the injection timing due to some factor. Factors that cause the continuous on mode include, for example, mechanical factors that cause the injector nozzle to be fixed while the valve is open, as well as electrical factors that are electrically driven in the valve opening direction due to an abnormality in the current-carrying path to the injector. There is.

  For example, in the case of a piezo injector, if an abnormality occurs such that the discharge path is disconnected after the valve is opened due to charging, the piezo element remains charged even after the injection timing ends, and therefore the piezo injector remains open. turn into. Also, in the case of a so-called solenoid injector that opens and closes by energizing the solenoid, for example, if an abnormality that energizes the solenoid constantly occurs, the solenoid injector remains open.

When the continuous on mode occurs, fuel may be excessively injected into the cylinder. Therefore, it is necessary to detect this and take appropriate measures.
On the other hand, in Patent Document 1, in a common rail fuel injection system, a fuel leak amount QL is calculated at each injection timing, and the fuel leak amount QL is calculated based on whether the calculated fuel leak amount QL is equal to or greater than a predetermined value Q1. Techniques for detecting outbreaks are described. The calculation of the fuel leakage amount QL in Patent Document 1 undergoes various calculations using the amount of fuel supply from the fuel supply pump, various leakage amounts, and changes in fuel pressure in the common rail (hereinafter also referred to as “common rail pressure”). Is done.

  When the continuous on mode occurs, the fuel is injected excessively than in the normal state. Therefore, if the technique described in Patent Document 1 is used, even when the continuous on mode occurs, the fuel is excessively injected by the continuous on mode, thereby detecting the occurrence of fuel leakage. The person can know the occurrence of abnormality.

JP 9-177586 A

  However, the technique described in Patent Document 1 calculates the fuel leakage amount QL by various calculations at each injection timing (for example, every rotation of the crank 180 degrees), and detects the occurrence of fuel leakage based on the fuel leakage amount QL. And detecting (determining) that fuel leakage has occurred when the number of times that the fuel leakage amount QL has been determined to be equal to or greater than the predetermined value Q1 has reached a predetermined number of times (for example, 10 times). is there.

  Therefore, it takes a considerable time from when the continuous on mode actually occurs until when an abnormality (fuel leakage occurrence) is detected. Therefore, with the technique described in Patent Document 1, it is difficult to detect the occurrence of abnormality at an early stage.

  That is, in the technique described in Patent Document 1, the engine rotates several times after the continuous on mode actually occurs until an abnormality is detected. Since the continuous on mode is an abnormal state in which fuel is continuously injected into the cylinder, it needs to be detected at an early stage, and it is too late to require several engine revolutions after the occurrence of the abnormality.

  If it is determined that the amount of fuel leakage QL is equal to or greater than the predetermined value Q1 even once, in order to detect it earlier, if it is determined that the fuel leakage has occurred immediately, it will be erroneously detected due to various factors such as characteristic variations of the injector and noise. May cause a decrease in detection accuracy.

  In addition, the technique described in Patent Document 1 can only detect the occurrence of fuel leakage and cannot distinguish whether it is due to the continuous on mode or due to other factors.

  As described above, the technique described in Patent Document 1 that detects the occurrence of fuel leakage based simply on the comparison result between the fuel leakage amount QL and the predetermined value Q1 has various problems as a method for detecting the continuous on mode. Is not appropriate.

  The present invention has been made in view of the above problems, and an object of the present invention is to enable accurate and rapid detection of an injector's continuous on mode (particularly, a continuous on mode generated by an electrical factor).

  The injector drive device of the present invention made to solve the above-described problems includes an injector control unit, an energized state abnormality determining unit, an abnormal discharge state determining unit, and an abnormality determining unit.

  The injector control means controls fuel injection from the injector by controlling energization to the injector to open or close the injector. The energization state abnormality determining means determines an abnormality in the energization state of the injector. The abnormal discharge state determination means determines whether or not the abnormal discharge state in which the fuel is abnormally discharged from the injector based on the change in the fuel pressure. The abnormality determining means holds the injector in a valve-opened state when it is determined by the energized state abnormality determining means that an abnormality in the energized state has occurred and when the abnormal discharge state determining means determines that the abnormal discharge state has occurred. It is determined that there is a continuous valve opening abnormal state.

  According to the injector drive device of the present invention configured as described above, it is determined that there is a continuous valve-opening abnormal state both when an abnormality occurs in the energized state and when it is in an abnormal discharge state. I have to. Moreover, the determination of the abnormal discharge state is realized by a simple method based on the change in the fuel pressure. Therefore, it is possible to detect a continuous valve opening abnormal state (continuous ON mode) accurately and quickly.

  In addition, the code | symbol in the parenthesis described in the claim shows the correspondence with the specific means as described in embodiment mentioned later as one aspect, Comprising: The technical scope of this invention is limited is not.

It is a block diagram showing schematic structure of the fuel-injection system of embodiment. It is a block diagram showing the internal structure of ECU. It is a time chart for demonstrating the drive control operation | movement of a piezo injector by ECU. It is a time chart for demonstrating the detection principle of the continuous on mode of a piezo injector by ECU. It is a flowchart showing a continuous on mode detection process. It is a time chart showing processing at the time of continuous on mode.

Preferred embodiments of the present invention will be described below with reference to the drawings.
(1) Overall Configuration of Fuel Injection System A fuel injection system 1 shown in FIG. 1 is for injecting fuel into, for example, a four-cylinder diesel engine (hereinafter also simply referred to as “engine”) 3 for automobiles. . The fuel injection system 1 includes a fuel supply pump 5, a common rail 10, a plurality (four in this embodiment) of piezo injectors (fuel injection valves) 20, and an ECU (Electronic Control Unit) 30. ing.

  The fuel supply pump 5 has a built-in feed pump that pumps fuel from the fuel tank 26. The fuel supply pump 5 is a known pump that pressurizes the fuel sucked from the feed pump into the pressurizing chamber when the plunger reciprocates as the camshaft cam rotates in the engine 3.

  A metering valve 7 is installed in the fuel supply pump 5, and the fuel intake amount (or the fuel discharge amount from each plunger) that each plunger of the fuel supply pump 5 sucks in the intake stroke is adjusted by the energization control by the ECU 30. To measure.

  The common rail 10 is a hollow member that forms a pressure accumulating chamber that accumulates fuel discharged from the fuel supply pump 5. The common rail 10 is provided with a pressure sensor 12 for detecting internal fuel pressure (common rail pressure PC) and an electromagnetically driven pressure reducing valve 14 for reducing the common rail pressure PC by energization control by the ECU 30.

  The engine 3 is provided with a crank angle sensor 24 that generates a crank angle signal indicating the rotation angle every time the crank of the engine 3 rotates by a predetermined angle as a sensor for detecting the driving state. The ECU 30 calculates the engine speed based on the crank angle signal from the crank angle sensor 24.

  Further, in the fuel injection system 1, although not shown as other sensors for detecting the driving state, an accelerator sensor for detecting an accelerator opening that is an operation amount of an accelerator pedal by a driver, a temperature of cooling water (water temperature) And various temperature sensors for detecting the temperature of intake air (intake air temperature).

  The four piezo injectors 20 are respectively installed in the cylinders 22 of the four-cylinder engine 3 and inject the fuel accumulated in the common rail 10 into the corresponding cylinders 22. Since the detailed configuration of the piezo injector 20 is already well known, a detailed description thereof is omitted here, but the outline thereof is as follows.

  That is, the piezo injector 20 includes a piezo stack Pa (see FIG. 2) in which a plurality of piezoelectric elements (piezo elements) are stacked, and the piezo stack Pa functions as an actuator by expanding and contracting due to the reverse piezoelectric effect. . That is, the piezo injector 20 includes a piezo actuator composed of a piezo stack Pa. The piezo stack Pa is a capacitive load that expands when charged and shrinks when discharged.

  When the piezo stack Pa is in a contracted state because the piezo stack Pa is not charged, a nozzle needle (not shown) included in the piezo injector 20 is closed to close the fuel injection port, and fuel is not injected. On the other hand, when the piezo stack Pa is extended by charging the piezo stack Pa, the nozzle needle is opened to open the fuel injection port, and the fuel supplied from the common rail 10 is injected from the fuel injection port.

In the following description, “charging” or “discharging” of the piezo injector 20 means charging or discharging of the piezo stack Pa.
(2) Configuration of ECU Next, the configuration and operation of the ECU 30 will be described with reference to FIG. As shown in FIG. 2, the ECU 30 mainly includes a DC / DC converter 41, a capacitor C0, a charge / discharge control circuit 42, a cylinder selection circuit 43, a control IC 44, and a microcomputer 45.

  The microcomputer 45 has a well-known configuration including a CPU, a RAM, a ROM, a flash memory, and the like, and executes various controls when the CPU executes a control program stored in the ROM or the flash memory.

  For example, the microcomputer 45 controls the energization amount to the metering valve 7 so that the common rail pressure PC detected by the pressure sensor 12 becomes the target pressure, and regulates the discharge amount of the fuel supply pump 5. Specifically, for example, the energization current value to the metering valve 7 is set based on a characteristic map representing the correlation between the energization current value to the metering valve 7 and the discharge amount. An energization current value for controlling the metering valve 7 is determined by, for example, a duty ratio, and a pump control signal indicating the duty ratio is output to the fuel supply pump 5.

  In addition, the microcomputer 45 sets the fuel injection amount and fuel injection timing of each piezo injector 20 and the pattern of multi-stage injection that performs pilot injection, pre-injection before pilot injection, after-injection, post-injection after pilot injection, etc. Control.

  Specifically, the microcomputer 45 calculates the required injection amount that satisfies the output torque requirement of the engine 3 based on the accelerator opening detected by the accelerator sensor, the crank angle detected by the crank angle sensor 24, and the like. Based on the calculated required injection amount, common rail pressure PC, and the like, an injection command value (command injection period) for the piezo injector 20 is calculated for each cylinder, and an injection signal for each cylinder corresponding to the injection command value (see FIG. 3 (a)) is output to the control IC 44.

  The injection signal output from the microcomputer 45 defines the charge start timing and the discharge start timing of the piezo stack Pa included in the piezo injector 20, and as shown in FIG. ) At the timing (time t02) at which the charging starts and falls to the L level. The drive time that is the length of the injection signal (the length of the H level state) is determined according to the required injection amount and the common rail pressure PC.

  The control IC 44 controls various circuits such as the charge / discharge control circuit 42 and the cylinder selection circuit 43 based on the injection signal input from the microcomputer 45, thereby controlling the charge / discharge of the piezo injector 20, and the fuel of the required injection amount. The piezo injector 20 is expanded and contracted so that is injected at a desired timing. Thereby, various injection modes, such as the injection quantity of the fuel injected from the piezo injector 20, the injection timing, and the injection stage, are controlled.

  The DC / DC converter 41 boosts the voltage of the battery Ba (for example, 12V) to a high voltage (for example, 150 to 300V) necessary for charging the piezo stack Pa, and the boosted high voltage is supplied to the capacitor C0. Output.

  The capacitor C0 accumulates electric charges for charging the piezo stack Pa while maintaining the high voltage from the DC / DC converter 41. One terminal of the capacitor C0 is connected to the DC / DC converter 41 side, and the other terminal is grounded via a resistor R1. The capacitance of the capacitor C0 is set to a relatively large value (for example, about several tens to several hundreds μF) such that the voltage hardly changes by a single charging process to the piezo stack. . The resistor R1 is for detecting the current flowing through the capacitor C0, and the voltage Vic across the resistor R1 is input to the control IC 44 as information indicating the current Ic flowing through the capacitor C0.

  The charging / discharging control circuit 42 is connected to two of the four piezo injectors 20 and controls charging / discharging of the two piezo injectors 20. In practice, two charge / discharge control circuits 42 are provided, and each control charge / discharge of two different piezo injectors 20. In FIG. 2, only one charge / discharge control circuit 42 is shown for simplicity of explanation, and only one of the two piezo injectors 20 controlled by the charge / discharge control circuit 42 is also used for the piezo injector 20. It is shown. In addition, bank selection switches are provided between the charge / discharge control circuits 42 and the two piezo injectors 20 corresponding thereto, but they are not shown in FIG.

  A cylinder selection circuit 43 is connected to the other end of each piezo stack Pa. Although the cylinder selection circuit 43 is actually provided for each of the four piezo stacks Pa, only one is shown in FIG.

  That is, the fuel injection system 1 of the present embodiment actually includes four piezo injectors 20, of which two piezo injectors 20 are set as one set, and a bank selection switch (not shown) is provided upstream of each energization path. ) To the corresponding charge / discharge control circuit 42. A cylinder selection circuit 43 is connected to the downstream side of the energization path of each piezo injector 20. However, in FIG. 2, only one piezo injector 20 is shown for simplification of explanation, and only one of the charge / discharge control circuit 42 and the cylinder selection circuit 43 connected to the one piezo injector 20 is also shown. Show. In the following description, charge / discharge control of one piezo injector 20 will be described as a representative.

  One terminal side of the capacitor C0, that is, the DC / DC converter 41 side is connected to one end of the piezo stack Pa (high potential) of the piezo injector 20 through a series connection body of the charging switch Tr1 and the charging / discharging coil L1. Terminal side). The other end of the piezo stack Pa (the terminal side that is at a low potential) is connected to the cylinder selection circuit 43 and grounded via the cylinder selection circuit 43.

  The cylinder selection circuit 43 includes a cylinder selection switch (n-channel MOSFET in this example) Tr3 having one end (drain) connected to the piezo stack Pa and the other end (source) connected to one end of the resistor R7, and one end of the cylinder selection circuit 43 The resistor R7 is connected to the other end (source) of the selection switch Tr3 and the other end is grounded. The resistor R7 is for detecting the current flowing through the piezo stack Pa, and the voltage Vip at one end of the resistor R7 is input to the control IC 44 as information indicating the current (piezo drive current Ip). A diode D3 is connected between the source and drain of the cylinder selection switch Tr3.

  In addition, one terminal (drain) of the discharge switch Tr2 is connected between the charge switch Tr1 and the charge / discharge coil L1, and the other terminal (source) of the discharge switch Tr2 is connected via the resistor R4. Grounded. The charge switch Tr1 and the discharge switch Tr2 are driven by a drive signal from the driver IC 46. The driver IC 46 drives the switches Tr1 and Tr2 by outputting a drive signal to the bases of the switches Tr1 and Tr2 in accordance with a control signal from the control IC 44. Note that the charge switch Tr1 and the discharge switch Tr2 are both n-channel MOSFETs in this embodiment.

  A diode D2 is connected in parallel to the discharge switch Tr2. The diode D2 has a cathode connected to the drain of the discharge switch Tr2, and an anode connected to the source of the discharge switch Tr2. The diode D2, together with the capacitor C0, the charge switch Tr1, and the charge / discharge coil L1, constitutes a chopper circuit that charges the piezo stack Pa, and functions as a freewheeling diode.

  On the other hand, a diode D1 is connected in parallel to the charging switch Tr1. The diode D1 has a cathode connected to the drain of the charge switch Tr1 and an anode connected to the source of the charge switch Tr1. The diode D1, together with the capacitor C0, the charge / discharge coil L1, and the discharge switch Tr2, constitutes a chopper circuit that discharges the electric charge of the piezo stack Pa, and functions as a freewheeling diode.

  Between the charge / discharge coil L1 and the piezo stack Pa, a diode D4 and a series connection body of a resistor R5 and a resistor R6 are connected in parallel to the piezo stack Pa, respectively. The diode D4 prevents the voltage of the piezo stack Pa from becoming negative. The voltage between the resistors R5 and R6 is input to the control IC 44 as information indicating the charging voltage (piezo voltage) Vp of the piezo stack Pa.

  In addition, a series connection body of a resistor R2 and a resistor R3 is connected to the capacitor C0 in parallel. The voltage at the connection point between the resistor R2 and the resistor R3 is input to the control IC 44 as information indicating the charging voltage of the capacitor C0.

  Further, the control IC 44 monitors the piezo voltage Vp, generates a signal (charge / discharge monitor signal) corresponding to the charge state of the piezo stack Pa, and outputs the signal to the microcomputer 45. The charge / discharge monitor signal will be described in detail later.

(3) Basic Operation of Piezo Injector Next, the operation of the piezo injector 20 will be described with reference to FIG. When the injection signal from the microcomputer 45 to the control IC 44 changes to H level (time t01 in FIG. 3A), the control IC 44 turns on the cylinder selection switch Tr3 and also performs chopper control by turning on / off the charging switch Tr1. To start. By this chopper control, a piezo driving current Ip (charging current) as shown in FIG. 3B flows through the piezo stack Pa, and as a result, the piezo voltage Vp gradually increases (FIG. 3C). . Further, as the piezo voltage Vp increases, the nozzle of the piezo injector 20 gradually opens as shown in FIG.

  Further, as shown in FIG. 3 (d), the charge / discharge monitor signal is at the H level when the piezo stack is not normally charged, but the current is supplied to the piezo stack Pa by the start of the chopper control of the charge switch Tr1. Starts to flow, the piezo drive current Ip starts to be energized and then goes to the L level.

  When a predetermined charge stop condition for satisfying the nozzle opening state of the piezo injector 20 is satisfied, the control IC 44 turns off the charge switch Tr1 and stops the chopper control. Various charging stop conditions are conceivable, for example, when the piezo voltage Vp reaches a predetermined target charging voltage or when the charging energy of the piezo stack Pa reaches a predetermined target charging energy. After the charge stop condition is satisfied and the chopper control of the charge switch Tr1 is stopped, the valve open state of the piezo injector 20 is maintained until time t02.

  Thereafter, when the injection signal from the microcomputer 45 to the control IC 44 changes to the L level at a time t02 when a predetermined drive time has elapsed from the start of charging, the control IC 44 starts chopper control by turning on / off the discharge switch Tr2.

  By this chopper control, a discharge current as shown in FIG. 3B flows through the piezo stack Pa, and as a result, the piezo voltage Vp gradually decreases (FIG. 3C). Further, as the piezo voltage Vp decreases, the nozzles of the piezo injector 20 are gradually closed as shown in FIG.

  When a predetermined discharge stop condition is satisfied, the control IC 44 turns off the discharge switch Tr2 and stops the chopper control. As for the discharge stop condition, various conditions can be considered as long as the discharge of the piezo stack Pa is sufficiently performed and the discharge can be stopped in a state where the piezo injector 20 is completely closed.

  Further, as shown in FIG. 3D, the charge / discharge monitor signal is generated when the piezo stack Pa starts to discharge at time t02 and then the piezo voltage Vp becomes equal to or lower than a predetermined threshold value Vth. (Time t03), the level changes to H level. In other words, the charge / discharge monitor signal changes to L level when current starts to flow through the piezo stack Pa (time t01), and after the discharge starts (time t02), the piezo voltage Vp falls below the threshold value Vth. Then (time t03), the signal turns to H level again.

  By the way, if the energization path from the ECU 30 to the piezo injector 20 is normal, the discharge of the piezo stack Pa proceeds when the discharge is started at time t02, and the charge / discharge monitor signal becomes H level at time t03.

  However, after the piezo injector 20 is opened due to charging, for example, the wiring from the ECU 30 to the piezo stack Pa is broken, for example, some electrical abnormality that prevents the charge charge of the piezo stack Pa from being discharged (abnormal current state). May occur (see “(disconnection)” in FIG. 3).

  When such an abnormality occurs, even if chopper control for discharging is started at time t02, the piezo voltage Vp does not decrease (FIG. 3 (c)), as shown by the one-dot chain line in FIG. The 20 nozzles remain open (FIG. 3E). If the valve opening state is maintained in this manner, an abnormal discharge state (continuous on mode) in which fuel is abnormally discharged from the piezo injector 20 is brought about.

  Therefore, in the present embodiment, the microcomputer 45 is configured to detect an abnormality (continuous on mode) in which the piezo stack Pa cannot be discharged as described above and the piezo injector 20 is held open. ing.

  In other words, if the discharge cannot be performed due to the occurrence of the continuous on mode, the piezo voltage Vp is maintained even if the control IC 44 starts chopper control for discharge (hereinafter also simply referred to as “discharge control”) due to the L level falling of the injection signal. It does not fall below the threshold value Vth. Therefore, the charge / discharge monitor signal output from the control IC 44 to the microcomputer 45 remains at the L level even after the discharge control is started.

  Further, the common rail pressure PC is feedback-controlled so as to be always a constant pressure. However, when the continuous on mode occurs and the valve is always opened, and the fuel is constantly injected, the feedback is performed. The followability of the control is lowered, and the common rail pressure PC is gradually lowered (see FIG. 4C).

  Therefore, if the charge / discharge monitor signal remains at the L level even after a predetermined time has elapsed after the start of the discharge control, the microcomputer 45 makes a temporary decision that the continuous on mode has occurred. Then, after making such a tentative judgment, based on the common rail pressure PC or the like, it is further determined whether or not abnormal fuel injection is being performed (calculation of the abnormal injection amount), and then continuously. Determine whether the on-mode has occurred.

  That is, in this embodiment, the microcomputer 45 can detect the continuous on mode with high accuracy and speed by looking at both the electrical information (charge / discharge monitor signal) and the system information (abnormal injection amount) in the fuel injection system 1. It is configured as follows.

(4) Continuous On Mode Detection Next, a specific detection procedure of the continuous on mode by the microcomputer 45 will be described in more detail with reference to FIG.

  The microcomputer 45 normally acquires the common rail pressure PC at a predetermined control cycle (for example, 1 msec), and performs the various controls described above based on the common rail pressure PC. When the injection start timing arrives, the microcomputer 45 raises the injection signal to the H level for a predetermined period as shown in FIG. During this period, as described with reference to FIG. 3, the piezo injector 20 is opened, and fuel injection is performed. Further, when energization (charging) to the piezo stack Pa is started at the rise of the injection signal at the H level, the charge / discharge monitor signal falls to the L level (FIG. 4B), which is also as described in FIG. It is.

  Then, after the start of injection, if an abnormality (continuous on mode) that prevents discharge from the piezo stack Pa to the ECU 30 occurs, for example, the wiring from the ECU 30 to the piezo injector 20 is disconnected, the injection signal becomes L level. Even after the predetermined time has elapsed after falling, the charging is completed and the charging / discharging monitoring signal should have changed to the H level (between times t4 and t5 in FIG. 4) has passed. The discharge monitor signal remains at the L level.

  Therefore, the microcomputer 45 confirms whether the charge / discharge monitor signal is at the H level at the first control timing (time t5 in the example of FIG. 4) after the predetermined period has elapsed after the injection signal is lowered to the L level. To do. Here, if the charge / discharge monitor signal is at the H level, the control is continued as usual assuming that the continuous on mode is not set.

  On the other hand, when the charge / discharge monitor signal remains at the L level at time t5 (when no rising edge has occurred), a temporary provisional determination that the continuous on mode has occurred is triggered by this. And proceed to a specific determination based on the system information (abnormal injection amount).

  The microcomputer 45 calculates the abnormal injection amount as follows. That is, the pressure deviation ΔPC (= PC (i), which is the difference between the common rail pressure PC (i−1) at the previous control timing (time t4) and the common rail pressure PC (i) at the current control timing (time t5). -1) -PC (i)) is calculated. Then, based on the calculated pressure deviation ΔPC, common rail pressure PC (i), various parameters such as bulk modulus, etc., the abnormal injection amount (1 msec) during one control cycle period (1 msec) from time t4 to time t5. (Corresponding to the difference from the injection amount assumed at normal time).

  The abnormal injection amount calculated in this way is always almost zero at the normal time when the continuous on mode is not generated. However, when the continuous on mode occurs due to disconnection or the like, the injection continues (abnormal injection amount occurs) even though the injection should stop if it is normal, and this causes the time of FIG. The common rail pressure PC gradually decreases as after t4. Therefore, in the present embodiment, the abnormal injection amount is calculated using the change in the common rail pressure PC.

  When the calculated abnormal injection amount is equal to or greater than a preset continuous on mode detection threshold Qth, it is determined that there is a high possibility that abnormal fuel injection is being performed and the continuous on mode is occurring. . That is, in the example at time t5 in FIG. 4, after an abnormality is detected based on electrical information (charging / discharging monitor signal), an abnormality determination is further performed based on system information (abnormal injection amount). Then, as shown in FIG. 4D, since the abnormal injection amount is equal to or greater than the on-mode detection threshold Qth, it is determined that the possibility of the continuous on-mode is high, and continuous as shown in FIG. The on-mode detection counter N is incremented from the initial value 0 to 1.

  Note that the determination that the continuous on mode is occurring at the time t5 may be performed, but in the present embodiment, the same determination is repeatedly performed for each control cycle, and the continuous on mode detection counter is determined. When N is equal to or greater than a predetermined threshold (continuous on-mode system condition confirmation margin) Nth, a determination is made that the continuous on-mode has occurred. The reason is roughly as follows.

  That is, since the abnormal injection amount (leakage amount) varies from injector to injector, it is necessary to set the continuous on-mode detection threshold Qth in consideration of the variation. The continuous on-mode detection threshold Qth needs to be set as high as possible in order to prevent erroneous detection of the continuous on-mode, but if it is set too high, it actually becomes abnormal (continuous on-mode). Despite this, it may not be detected. Conversely, if the continuous on-mode detection threshold Qth is set low to prevent detection omission, there is a possibility that an abnormality is erroneously detected even though it is not actually abnormal (continuous on mode).

  In addition, there may be a case where the amount of decrease in the common rail pressure PC is still small due to the operation of the fuel supply pump 5 in spite of actually being in the continuous on mode. In such a case, the calculation of the abnormal injection amount is possible. The result may also be a small value and erroneously determined that no abnormality has occurred.

  Therefore, in this embodiment, the continuous on-mode detection threshold Qth is set to a rather low value to prevent detection omission. In addition, in order to prevent erroneous detection, the continuous on mode is not determined by only one abnormality detection, but a margin is provided for the number of detections, and abnormality detection is performed a predetermined number of times or more (in this example, Nth times or more). The continuous on mode is determined for the first time.

  However, setting the continuous on-mode detection threshold Qth to a lower value is merely an example, and conversely, it may be set to a higher value to prevent erroneous detection. The appropriate value can be determined as appropriate.

  Note that it is not essential that the threshold value is equal to or greater than the continuous on-mode detection threshold Qth for Nth times. The continuous on-mode detection counter N does not count the number of times that the abnormal injection amount becomes equal to or greater than the threshold value Qth continuously every control timing, but cumulatively adds the number of times that the abnormal injection amount becomes equal to or greater than the threshold value Qth. However, when the charge / discharge monitor signal becomes H level, the continuous on-mode detection counter N is cleared to zero.

  In the example of FIG. 4, when the continuous on-mode detection counter N is incremented to 1 by detecting an abnormality in both the electrical information and the system information at time t5, the same applies at time t6 that is the next control timing. Judgment is made.

  That is, the microcomputer 45 first makes an abnormality determination based on electrical information (charge / discharge monitor signal) even at time t6. At this time, if the charge / discharge monitor signal remains at the L level, a temporary determination is made that the continuous on mode is occurring as in the case of time t5, and the specific based on the system information (abnormal injection amount). Proceed to manual judgment. The calculation of the abnormal injection amount at time t6 is also the pressure that is the difference between the common rail pressure PC (i) at the previous control timing (time t5) and the common rail pressure PC (i + 1) at the current control timing (time t6). Deviation ΔPC (= PC (i) −PC (i + 1)) is calculated. Then, based on the calculated pressure deviation ΔPC, various parameters such as the common rail pressure PC, the bulk modulus, etc., the abnormal injection amount during one control cycle period (1 msec) from time t5 to time t6 Is calculated.

  Then, it is determined whether or not the calculated abnormal injection amount is equal to or greater than the continuous on-mode detection threshold Qth. If it is equal to or greater than the continuous on-mode detection threshold Qth, the continuous mode detection counter N is incremented by one. In the example of FIG. 4, since the abnormal injection amount calculated at time t6 is also equal to or greater than the continuous on mode detection threshold Qth, the continuous on mode detection counter N is incremented to 2.

  If the abnormal injection amount is lower than the continuous on-mode detection threshold Qth at time t6, the continuous on-mode detection counter N is not changed, and abnormality determination is similarly performed after time t7. If the charge / discharge monitor signal rises to H level at time t6, the abnormal injection amount is not calculated and the continuous on-mode detection counter N is reset to zero. That is, unless there is an abnormality in the abnormality determination based on the electrical information, the calculation of the abnormal injection amount is not performed. In this case, the process waits until time t7, which is the next control timing, and again performs abnormality determination based on electrical information (charge / discharge monitor signal) at time t7. However, if the charge / discharge monitor signal rises to H level at time t6, the abnormality determination may not be performed until the next injection (that is, until the injection signal rises next time).

  After time t7, the abnormality determination is performed in the same manner for each control timing. That is, first, abnormality determination based on electrical information (charge / discharge monitor signal) is performed, and if it is determined that there is an abnormality based on electrical information, then abnormality determination is performed based on system information (abnormal injection amount).

  As a result of performing the abnormality determination at each control timing as described above, when the continuous on-mode detection counter N becomes Nth times or more, it is determined that the continuous on-mode has occurred. In the example of FIG. 4, abnormality determination based on electrical information and abnormality determination based on system information are made from time t5 to time t9, whereby the continuous on-mode detection counter N is set to the continuous on-mode system condition margin Nth ( In this example, Nth = 5) or more.

  When the continuous on-mode detection counter N becomes equal to or greater than the continuous on-mode system condition margin Nth, the continuous on-mode confirmation flag is set to 1 (H level) as shown in FIG. A final decision has been made.

  After the continuous on mode determination flag is set to the H level, measures such as forcibly stopping the fuel supply pump 5 and reducing the common rail pressure PC by the pressure reducing valve 14 are performed as described later. Further, after the continuous on mode determination flag is set to the H level, the abnormality determination is not performed for each control cycle. In the example of FIG. 4, it is shown that the abnormality determination is not performed after time t10. However, even after the continuous on mode is confirmed, the abnormality determination based on the electrical information may be continued. In that case, after both the electrical information and the system information become normal after the continuous on mode is determined, the continuous on mode determination flag may be cleared to L level and the fuel injection control may be resumed.

  On the other hand, unless the continuous on mode is determined, the abnormality determination based on the electrical information for each control timing is continued until the injection signal rises to the H level again (that is, until the next fuel injection is started). However, if the continuous on mode is not determined even after a certain period (for example, 10 msec) has elapsed, the abnormality determination may be stopped.

  Note that a control cycle (1 msec in this example), which is a time interval for performing abnormality determination, is much shorter than a time required for the crankshaft to rotate 180 degrees at a normal traveling speed (for example, 60 km / h). Therefore, it is possible to determine whether the continuous on mode is confirmed in a much shorter time than when the crankshaft rotates 180 degrees.

(5) Description of Control Process by Microcomputer Next, the continuous on mode detection process executed by the microcomputer 45 in order to determine the presence or absence of the continuous on mode described with reference to FIG. 4 will be described with reference to the flowchart of FIG. Every time the microcomputer 45 lowers the injection signal from the H level to the L level, the microcomputer 45 continues to raise the injection signal to the H level again from the control timing (time t5 in FIG. 4) after a certain period of time has elapsed since the falling. 4 is repeatedly executed at each control timing. However, after the continuous on mode determination flag is set in S170, this continuous on mode detection processing is not performed until the next fuel injection is performed.

  When the microcomputer 45 starts the continuous on-mode detection process of FIG. 5, first, in S110, the microcomputer 45 performs abnormality detection based on the charge / discharge monitor signal (electrical information). That is, it is determined whether the charge / discharge monitor signal has risen to the H level. If it is not abnormal (that is, if the charge / discharge monitor signal is H level), the process proceeds to S180, and the continuous on-mode detection counter N is cleared to zero.

  If it is determined in S110 that the charge / discharge monitor signal remains in the abnormal state, the pressure deviation ΔPC is calculated in S120, and the abnormal injection amount is calculated in S130. The specific methods of calculating the pressure deviation ΔPC in S120 and calculating the abnormal injection amount in S130 are as described above.

  In S140, it is determined whether or not the abnormal injection amount calculated in S130 is equal to or greater than the continuous on-mode detection threshold Qth. Here, when the abnormal injection amount is smaller than the continuous on-mode detection threshold Qth, the continuous on-mode detection process is terminated, but when the abnormal injection amount is equal to or greater than the continuous on-mode detection threshold Qth, in S150, the continuous The on-mode detection counter N is incremented by one.

  In S160, it is determined whether the continuous on-mode detection counter N is equal to or greater than the continuous on-mode system condition determination margin Nth. If the continuous on-mode detection counter N is smaller than the continuous on-mode system condition determination margin Nth, the continuous on-mode detection process is terminated, but the continuous on-mode detection counter N is equal to or greater than the continuous on-mode system condition determination margin Nth. If it is, in S170, the continuous on mode determination flag is set to 1 (H level), and this process is terminated.

  On the other hand, in parallel with the continuous on-mode detection process in FIG. 5, the microcomputer 45 also executes the continuous on-mode response process in FIG. 6 for each control cycle. When the microcomputer 45 starts the processing for responding to the continuous on mode in FIG. 6, first, in S210, the microcomputer 45 determines whether or not the continuous on mode determination flag is set to 1 (H level). If the continuous on-mode determination flag is not set to 1, the corresponding processing at the time of the continuous on-mode is terminated as it is, but if the continuous on-mode determination flag is set to 1, the injection stop processing is performed in S220. Do. Specifically, all the output of the injection signal of each cylinder is stopped (fixed at L level).

  In subsequent S230, a pump stop process is performed. Specifically, by stopping the operation of the fuel supply pump 5, the fuel supply (pressure feeding) from the fuel supply pump 5 to the common rail 10 is stopped.

Further, in S240, a pressure reducing valve opening process is performed. Specifically, the pressure reducing valve 14 is opened to forcibly reduce the common rail pressure PC of the common rail 10.
(6) Effects of Embodiment As described above, in the fuel injection system 1 of the present embodiment, the continuous on mode is detected based on both electrical information (charge / discharge monitor signal) and system information (abnormal injection amount). I am doing so.

  For example, it is not impossible to make an abnormality determination based only on the abnormal injection amount. However, if the abnormality determination is performed based on the abnormal injection amount, it is necessary to repeatedly perform various arithmetic processes such as calculation of the pressure deviation ΔPC and calculation of the abnormal injection amount based on the pressure deviation ΔPC. Will become very large. Moreover, even if it is determined that the abnormality is based only on the basis of the abnormal injection amount, it is not possible to distinguish whether the cause is due to the continuous on mode or other factors (for example, damage to the piping, etc.). It is extremely difficult.

  The same applies to electrical information, and it is not impossible to make an abnormality determination based only on electrical information. However, even in the case of abnormality determination based only on electrical information, it is difficult to determine whether or not it is actually in the continuous on mode even if it is determined to be abnormal. For example, after the charge to the piezo stack Pa is started, if the discharge path is disconnected while the charge voltage value during the charge is still low, the charge voltage is low and the piezo injector 20 does not open. For this reason, in this case, the charge state of the piezo stack Pa is maintained, but the continuous on mode is not established. However, in the abnormality determination based only on the electrical information, even if the continuous on mode is not set, the electrical information (charging / discharging monitor signal) is erroneously determined as the continuous on mode because the electrical information (charge / discharge monitor signal) does not rise to the H level. . Therefore, abnormality determination based only on electrical information is insufficient to accurately determine whether or not the continuous on mode is set.

  Therefore, in this embodiment, first, the presence / absence of an abnormality is determined based on the charge / discharge monitor signal, and when an abnormality is determined based on the charge / discharge monitor signal, an abnormality determination based on the abnormal injection amount is further performed. Moreover, the calculation of the abnormal injection amount can be performed quickly and easily using the pressure deviation ΔPC and the like, and can be sufficiently performed even in a short control cycle. By adopting such a configuration, it is usually only necessary to check the level of the charge / discharge monitor signal, and an abnormality determination based on the abnormal injection amount is not performed unless the continuous on mode occurs. A significant reduction is realized.

  Then, the final abnormality determination is performed based on both the charge / discharge monitor signal and the abnormal injection amount while greatly reducing the software processing load. Specifically, when an abnormality is detected by the charge / discharge monitor signal, the calculation of the abnormal injection amount and the determination of the abnormality are performed immediately. Therefore, it is possible to accurately and quickly determine the occurrence of the continuous on mode.

  Also, in this embodiment, after an abnormality is determined in both electrical information and system information at a certain control timing, the electrical information is checked again at each control timing until the continuous on mode is determined. Yes. While the abnormality is determined based on the electrical information, the number of times of abnormality determination (continuous on mode detection counter N) is incremented each time an abnormality is determined based on the system information, and the number of times of abnormality determination is a predetermined value (continuous on mode). When the system condition determination margin (Nth) is exceeded, a determination is made that the continuous on mode has occurred. Therefore, the continuous on mode can be detected with higher accuracy.

  In this embodiment, an abnormality is determined with a control cycle that is much shorter than the rotation speed of the crankshaft. If an abnormality is detected by a charge / discharge monitor signal at a certain control timing, an abnormal injection is immediately performed at that control timing. Anomaly judgment based on quantity is also made. Therefore, even if the continuous on mode occurs after the start of fuel injection, the continuous on mode is determined and determined before the engine rotates once, and various countermeasures such as stopping the fuel supply pump 5 and opening the pressure reducing valve are quickly performed. Is called.

Therefore, even if the continuous on mode occurs, it can be detected quickly and various processes for stopping fuel injection can be performed quickly.
[Modification]
Although the embodiments of the present invention have been described above, the embodiments of the present invention are not limited to the above-described embodiments, and can take various forms as long as they belong to the technical scope of the present invention. Needless to say.

  For example, in the above embodiment, the charge / discharge monitor signal is used as the electrical information, but this is an example, and other signals may be used. That is, as long as the state of charge of the piezo stack Pa can be known directly or indirectly, there is no particular limitation on what signal is used as the electrical information.

  The same applies to the system information. In the above embodiment, the abnormal injection amount is used as the system information, but this is also an example. That is, as long as it can be directly or indirectly known from the piezo injector 20 that abnormal fuel injection is being performed, there is no particular limitation on what signal is used as system information. .

In addition, as a specific method for calculating the abnormal injection amount, the method using the pressure deviation ΔPC is exemplified in the above embodiment, but this is only an example.
Further, the present invention can be applied not only to a piezo injector but also to a solenoid injector. In the case of a solenoid injector, for example, when the power supply is short-circuited on the high side and the ground is short-circuited on the low side, there is a possibility that the energization is always performed regardless of the injection signal and the continuous on mode is set. Therefore, by using a signal that can detect such an abnormality (that energization continues despite the injection signal being turned off) as electrical information, the detection / determination determination of the continuous on mode is performed as described above. Can do.

  Furthermore, the application of the present invention is not limited to application to a common rail fuel injection system. The present invention can be applied to any fuel injection system having any structure that may cause an abnormality in which the valve-opening state continues due to the continuous on-mode occurrence, thereby abnormally injecting fuel into the cylinder.

  DESCRIPTION OF SYMBOLS 1 ... Fuel injection system, 3 ... Engine, 5 ... Fuel supply pump, 7 ... Metering valve, 10 ... Common rail, 12 ... Pressure sensor, 14 ... Pressure reducing valve, 20 ... Piezo injector, 22 ... Cylinder, 24 ... Crank angle sensor , 26 ... Fuel tank, 30 ... ECU, 41 ... DC / DC converter, 42 ... Charge / discharge control circuit, 43 ... Cylinder selection circuit, 44 ... Control IC, 45 ... Microcomputer, 46 ... Driver IC, Ba ... Battery, C0 ... Capacitors, D1 to D4 ... Diodes, L1 ... Charge / discharge coils, Pa ... Piezo stacks, R1 to R7 ... Resistances, Tr1 ... Charge switches, Tr2 ... Discharge switches, Tr3 ... Cylinder selection switches

Claims (4)

Injector control means (44, 45) for controlling fuel injection from the injector by controlling energization to the injector to open or close the injector;
Energization state abnormality determining means (44, 45, S110) for determining abnormality of the energization state of the injector;
Abnormal discharge state determination means (45, S140) for determining whether or not the fuel is abnormally discharged from the injector based on a change in fuel pressure;
When the energized state abnormality determining means determines that the energized state abnormality has occurred and the abnormal discharge state determining means determines that the abnormal discharge state is present, the injector is held in the valve open state. Abnormality determining means (45, S170) for determining that the continuous valve opening abnormal state is performed;
An injector driving device comprising: The injector drive device according to claim 1,
The abnormal discharge state determination unit determines whether or not the abnormal discharge state is present when the energization state abnormality determination unit determines that an abnormality in the energization state has occurred. Injector drive device. The injector drive device according to claim 2,
The energized state abnormality determining means repeatedly executes the determination as to whether or not the energized state abnormality has occurred at every predetermined determination timing,
The abnormal discharge state determination means executes a determination as to whether or not the abnormal discharge state occurs each time it is determined by the energization state abnormality determination means that the abnormality in the energization state has occurred at the determination timing. ,
The abnormality determining means (S150 to S170) counts the determined number of times each time the abnormal discharge state is determined by the abnormal discharge state determining means, and the number of times becomes equal to or greater than a preset abnormality determination number threshold. Injector drive device characterized in that it is determined that the continuous valve-opening abnormal state has occurred. The injector drive device according to any one of claims 1 to 3,
The injector is a piezo injector (20) configured to realize valve opening or closing by extending or reducing the piezo element by charging or discharging the piezo element;
The injector control means opens the injector by starting charging of the piezo element at a predetermined charging start timing, and starts discharging of the piezo element at a predetermined discharge start timing after the start of the charging. And is configured to close the injector.
The energization state abnormality determining means monitors the charge voltage of the piezo element after a predetermined time has elapsed from the discharge start timing, and the energization state is determined when the charge voltage is not below a predetermined charge voltage drop threshold. An injector driving device characterized in that it is determined that an abnormal state has occurred.