JP2019132186A - Fuel injection control device - Google Patents

Abstract

To provide a fuel injection control device which can accurately valve-open and control an injector.SOLUTION: A charge control circuit 17 charges and controls a piezoelectric element 4 by discharging an electric charge accumulated in a charge capacitor 7 to the piezoelectric element 4 on the basis of a charge-on pulse P which is inputted at fuel injection, and a charging electric charge command value Qdc. The charge control circuit 17 comprises a correction calculator 26 for correcting the charging electric charge command value which is inputted at post-injection on the basis of a result in the past when an electric charge is charged to the piezoelectric element 4. The correction calculation part 26 corrects and calculates the charging electric charge command value which is inputted at the post-injection by using a total sum of a peak current value of a current flowing in the piezoelectric element 4 when the charge control circuit 17 charges and controls the electric charge to the piezoelectric element 4 in response to the charge-on pulse P, and the charging electric charge command value at that time as a result in the past.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a fuel injection control device that controls injection of fuel.

  In general, an injector using a capacitive actuator such as a piezo element is provided. The applicant has been developing a fuel injection control device that controls the injection of fuel using this injector. This type of fuel injection control device controls the opening and closing of an injector by controlling charging of a capacitive actuator based on a charge pulse and a charge charge command value input during fuel injection. Then, the charge is temporarily stored in the charge storage unit, and the fuel is injected from the injector by charging the stored electric power to the capacitive actuator. For example, Patent Document 1 describes a fuel injection control device that performs fuel injection control with an injector using a piezoelectric element.

  In the technology described in Patent Document 1, in order to appropriately suppress vibration when the piezo element is displaced, switching when increasing or decreasing the amount of current flowing through the piezo element is controlled for charge control and discharge control of the piezo element. It is done on the basis of becoming.

JP 2006-166649 A

  The inventor considers controlling the charge charged from the charge storage unit to the capacitive actuator to be constant in order to accurately perform the valve opening control process of the injector. In general, the charging energy E of the capacitive actuator can be calculated by subtracting the energy loss of the circuit used during charging from the released energy of the charge storage unit. That is, if the DC voltage stored in the charge storage unit is Vdc, the charge stored in the charge storage unit is Qdc, and the energy loss of the circuit is PL, the charging energy E of the piezo element can be calculated by E = Vdc × Qdc−PL. .

  The inventor assumed that the DC voltage Vdc and the circuit loss PL were constant and considered that the discharge charge Qdc was controlled to be constant, but various variations due to individual variations of injectors and variations of circuit constants were observed. It has been found that the variation of the circuit loss PL cannot be ignored if considered.

As a result of searching for the cause of the variation in the circuit loss PL, the inventor has found that the circuit loss PL changes substantially in proportion to the circuit emission energy and changes in proportion to the capacitance value Cinj of the capacitive actuator. ing. Therefore, the fluctuation of the capacitance value of the capacitive actuator also greatly affects the fluctuation of the circuit loss PL. However, since it is difficult to directly detect the circuit loss PL and the change in the capacitance value of the capacitive actuator, the inventor considers using alternative means.
An object of the present invention is to provide a fuel injection control device capable of accurately controlling the opening of an injector.

  The first aspect of the present invention is directed to a fuel injection control device that controls fuel injection by an injector using a capacitive actuator. The charge control unit generates a charge on pulse at the time of fuel injection based on the input charge charge command value, and discharges the charge accumulated in the charge accumulation unit to the capacitive actuator, thereby controlling the charge on the capacitive actuator. The charge control unit includes a correction arithmetic unit that corrects a charge charge command value input at the time of post-injection based on a past result of charging the capacitive actuator with charge. This correction computing unit is input at the time of post-injection using the peak current value of the current flowing through the capacitive actuator as a past result when the charge control unit performs charge control on the capacitive actuator in response to the charge on pulse. The charge charge command value is corrected. As a result, it is possible to accurately control the opening of the injector without directly detecting the circuit loss and the fluctuation of the capacitance value of the capacitive actuator.

The block diagram which shows the electric constitution of the injection control system in 1st Embodiment. Timing chart that outlines the overall flow Correlation diagram of total injector peak current and circuit loss Explanatory drawing of correction calculation processing Flow chart showing the flow of processing Explanatory diagram for estimating the N + 2th peak current sum from the Nth and N + 1 charge charge command values and the sum of injector peak currents Timing chart showing how to calculate the sum of peak currents corresponding to charge on pulse

  Hereinafter, an embodiment of a fuel injection control device will be described with reference to the drawings. FIG. 1 schematically shows an injection control system 1 for an internal combustion engine.

  A fuel injection control device 2 of the injection control system 1 drives N piezo injectors 3 that inject and supply fuel to an internal combustion engine such as a diesel engine mounted on a vehicle such as an automobile. The piezo injector 3 is connected to a common rail (not shown) in which high pressure fuel is accumulated. The fuel injection control device 2 controls the opening of the injection nozzle of the piezo injector 3 to inject fuel from the common rail to the internal combustion engine (not shown) through the injection nozzle of the piezo injector 3. The piezo injector 3 includes a piezo element 4 as a capacitive actuator, and driving / non-driving of the piezo element 4 is performed by charge and discharge.

  A fuel injection control device 2 shown in FIG. 1 is configured to charge and discharge electric charges in the piezo element 4 of the piezo injector 3, and controls the fuel injection timing. For simplification of description, the piezo injector 3 for one cylinder will be shown and described, but the contents of the present disclosure are similarly applied to the case of controlling piezo injectors of a plurality of cylinders such as two cylinders, four cylinders, and six cylinders. Applicable.

  The fuel injection control device 2 includes a microcomputer 5, a piezo injector control circuit (hereinafter abbreviated as a control circuit) 6, a charge capacitor 7 as a charge storage unit, a charge switch (also referred to as a switching element) 8, a discharge switch (also referred to as a switching element). ) 9, a charge / discharge coil 10, a cylinder selection switch (also referred to as a switching element) 11, voltage buffers 12 to 14, and a current detection resistor 15 are connected in the illustrated form.

  The microcomputer 5 includes, for example, a CPU, a ROM, a RAM, an I / O, and the like (all not shown), inputs various sensor signals (not shown), generates an injection command signal, and outputs the injection command signal to the control circuit 6. The microcomputer 5 is connected to the control circuit 6 so as to be capable of serial communication, and outputs a charge charge command value Qdc calculated based on various sensor signals to the control circuit 6 at the time of injection.

  The control circuit 6 is an integrated circuit device using, for example, an ASIC (Application Specific Integrated Circuit), and includes a control body such as a logic circuit and a CPU, and a storage unit 6a, and executes various controls based on hardware and software. The control is executed in response to various command signals from the microcomputer 5. The storage unit 6a includes registers such as a plurality of injection information registers 6aa (see FIG. 4 described later), and various memories such as RAM, ROM, and EEPROM. If described functionally, the control circuit 6 is divided into a power supply circuit 16, a charge control circuit (equivalent to a charge control unit) 17, a discharge control circuit 18, and a cylinder selection control circuit 19, as shown in FIG. ing.

  The piezo element 4 of the piezo injector 3 is connected between the upstream terminal 2 a and the downstream terminal 2 b of the fuel injection control device 2. A cylinder selection switch 11 and a current detection resistor 15 are connected in series between the downstream terminal 2 b and the ground node G. The cylinder selection switch 11 is composed of, for example, an N channel type MOS transistor with a body diode 11a. The cylinder selection control circuit 19 of the control circuit 6 applies a cylinder selection control signal C1 to the cylinder selection switch 11 via the voltage buffer 14 to turn on / off the cylinder selection switch 11.

  The power supply circuit 16 of the control circuit 6 boosts a battery voltage (not shown) (for example, several tens of volts) to a predetermined range (for example, about 200 V), and charges the charging capacitor 7 with the boosted voltage. The charging capacitor 7 is configured using, for example, an aluminum electrolytic capacitor. A charging switch 8 is connected to the subsequent stage of the charging capacitor 7. The charge switch 8 is constituted by, for example, an N channel type MOS transistor with a body diode 8a. A charge / discharge coil 10 is connected between the charge switch 8 and the upstream terminal 2a. The charging switch 8 is configured such that its body diode 8a is forward from the common connection point with the charging / discharging coil 10 toward the charging node N1 of the charging capacitor 7. The charge control circuit 17 of the control circuit 6 controls the on / off of the charge switch 8 by applying a charge switch control signal C2 to the charge switch 8 via the voltage buffer 12.

  A discharge switch 9 is connected between the node N2 and the ground node G that are commonly connected between the charge switch 8 and the charge / discharge coil 10. The discharge switch 9 is composed of, for example, an N channel type MOS transistor with a body diode 9a.

  The discharge switch 9 is configured such that its body diode 9a is in the forward direction from the ground node G toward the node N2. The discharge control circuit 18 of the control circuit 6 detects a current flowing through the current detection resistor 15 and applies a discharge switch control signal C3 to the discharge switch 9 via the voltage buffer 13 based on the detected current, thereby discharging the switch. 9 is turned on / off.

  As shown in FIG. 1, the charge control circuit 17 includes a charge-on control circuit 20, a logic gate 21, a charge measurement circuit 23, a threshold setting circuit 24, a comparison unit 25, a correction calculator 26, a peak calculation unit 27, and an update. A functional block as the unit 28 is provided. The charging control circuit 17 controls the charging of the piezo element 4 using various information stored in the storage unit 6a.

  When the injection command signal is input as the active level “H” from the microcomputer 5, the charge-on control circuit 20 generates a charge-on pulse P a predetermined number of times (for example, 14 times) at the time of fuel injection, and sequentially outputs these to the logic gate 21. . The logic gate 21 is composed of, for example, an AND gate, and performs a logical operation between the output of the charge-on control circuit 20 and the output of the comparison unit 25. Then, the logic gate 21 outputs this logic operation result as a charge switch control signal C2 to the control terminal of the charge switch 8, that is, the gate of the MOS transistor via the voltage buffer 12.

  When the charge measurement circuit 23 detects the current flowing through the current detection resistor 15 and inputs the charge measurement permission signal EN accompanying the charge on pulse P of the charge on control circuit 20, the charge measurement circuit 23 integrates the detected current with time from this input timing. Then, the charge measuring circuit 23 can measure the amount of charge accumulated in the piezo element 4 as the measured charge Qinj_a (“a” is the number of charge on pulses P). Then, the charge measurement circuit 23 outputs a voltage corresponding to the measurement charge Qinj_a to the comparison unit 25.

  The threshold setting circuit 24 sets a threshold voltage corresponding to the charge charge threshold Qth and outputs the threshold voltage to the comparison unit 25. This charging charge threshold value Qth is a threshold value provided corresponding to the measured charge Qinj_a obtained by integrating the charging current of the piezo element 4. The comparison unit 25 is configured by a comparator, for example, and compares the voltage corresponding to the measured charge Qinj_a measured by the charge measurement circuit 23 with the threshold voltage corresponding to the charge charge threshold value Qth of the threshold setting circuit 24, and the comparison result Is output to the logic gate 21. If the measured charge Qinj_a has not reached the charge charge threshold value Qth, the logic gate 21 continues to turn on the charge switch 8 by outputting the charge switch control signal C2 based on the charge on pulse P to the charge switch 8. When the measured charge Qinj_a reaches the charge charge threshold value Qth, the logic gate 21 controls the charge switch 8 to be turned off by outputting an off control signal to the charge switch 8 as the charge switch control signal C2. In this way, a feedback system is configured.

  Further, when the charge switch control signal C2 output from the logic gate 21 is input as the charge off signal, the peak calculation unit 27 acquires the current flowing through the current detection resistor 15 at the timing as the peak current value Ipeak. The update unit 28 is configured to update the stored content of the injection information register 6aa of the storage unit 6a when the charge off signal of the charge on pulse P is input.

<Basic operation explanation>
Here, the operation when the piezo injector 3 is controlled to open and close to inject fuel will be described. As usual, when the power supply based on the battery voltage is turned on to the fuel injection control device 2, the microcomputer 5 and the control circuit 6 cooperate to perform fuel injection control processing. When the fuel injection control device 2 injects fuel from the piezo injector 3, the microcomputer 5 outputs an active level “H” of an injection command signal corresponding to the piezo injector 3 to the control circuit 6 at timing t 1 in FIG. 2. When the control circuit 6 receives the active level “H” of the injection command signal, the cylinder selection control circuit 19 turns on the cylinder selection switch 11. Thereafter, the charge-on control circuit 20 generates and outputs the charge-on pulse P a plurality of times, thereby starting chopper control for turning on / off the charge switch 8. See period T1 in FIG.

  When the control circuit 6 performs chopper control, the piezoelectric drive current is repeatedly applied to the piezoelectric element 4 in a pulse shape. As a result, the voltage between the terminals of the piezo element 4 gradually increases. As the voltage between the terminals of the piezo element 4 gradually increases, the nozzles of the piezo injector 3 are gradually opened and the nozzles are opened to the maximum position. When a predetermined charge stop condition is satisfied, the control circuit 6 stops the chopper control and holds the charge switch 8 in the off state. The control circuit 6 holds the state where the charging switch 8 is turned off while the cylinder selection switch 11 is kept on until the injection command signal is received as the non-active level “L”. Thereby, fuel is injected into the internal combustion engine.

  Thereafter, when the microcomputer 5 outputs the injection command signal to the control circuit 6 as the non-active level “L” at the timing t2 in FIG. 2, the control circuit 6 receives the non-active level of the injection command signal. Then, the control circuit 6 starts chopper control by turning on / off the discharge switch 9 by the discharge control circuit 18. The control circuit 6 discharges the electric charge accumulated in the piezo element 4 by performing chopper control. At this time, the discharge control circuit 18 of the control circuit 6 repeatedly applies an on / off control signal in a pulse shape to the control terminal of the discharge switch 9 in the period T2 of FIG. For this reason, the electric charge accumulated in the piezo element 4 is discharged in pulses, and the nozzles of the piezo injector 3 are gradually closed.

  When a predetermined discharge stop condition is satisfied, the control circuit 6 stops the chopper control by turning off the discharge switch 9. At the same time, the cylinder selection switch 11 is stopped. Thereby, the fuel injection process can be stopped. As shown in FIG. 2, one injection process F1 is performed in such a flow. Thereafter, every time the control circuit 6 receives an injection command signal from the piezo injector 3, as shown in injection processes F2 to F5. The process will be repeated.

<Significance of correction processing based on peak current value Ipeak>
In the present embodiment, during such charge control, the correction calculator 26 sums the peak currents ΣIpeak that flow into the piezo elements 4 corresponding to past charge results, in particular, past charge on-pulses P, in which the charge is charged. Is used to correct the charge charge command value Qdc_x input from the microcomputer 5 and calculate the corrected charge charge command value (hereinafter referred to as “charge charge correction command value Qdc_x_adjust”). Said x is a number which shows the order of an injection command signal. For x, among N, N + 1, and N + 2 that frequently appear in the following description, N + 2 is the number of the current injection command signal for convenience. Further, N and N + 1 are the numbers of the injection command signals two times before and once before the current time, respectively. Hereinafter, first, the significance of the correction process using the sum ΣIpeak of the peak current will be described.

It is desirable that the fuel injection control device 2 controls the electric charge released from the charging capacitor 7 to be constant in order to accurately perform the valve opening control process of the piezo injector 3. The charging energy E of the piezo element 4 can be calculated by subtracting the loss of the circuit used at the time of discharging from the discharging energy of the charging capacitor 7. That is, if the DC voltage stored in the charging capacitor 7 is Vdc, the charge discharged from the charging capacitor 7 is Qdc, and the circuit loss is PL, the charging energy E of the piezo element 4 can be calculated by the following equation (2). .
E = Vdc × Qdc-PL (2)

  Conventionally, the discharge charge Qdc is controlled to be constant on the assumption that the DC voltage Vdc and the circuit loss PL are constant. However, if various variations due to individual variations of injectors and variations of circuit constants are taken into consideration, The variation of the circuit loss PL cannot be ignored.

  The breakdown of the circuit loss PL of the fuel injection control device 2 includes the energization loss and switching loss of the switching elements (charge switch 8, discharge switch 9, cylinder selection switch 11) excluding the internal power loss of the microcomputer 5 and the control circuit 6. It can be divided into flywheel energization loss of the body diodes 8a, 9a and 11a, and circuit power loss of other voltage buffers 12 to 14 and the like. Among these various circuit losses PL, it has been found that the ratio of the switching loss of the switching elements (the charge switch 8, the discharge switch 9, and the cylinder selection switch 11) occupies most. For this reason, reducing this switching loss variation leads to a significant reduction in the variation in circuit loss PL.

  It is impossible to directly detect this switching loss while the fuel injection control device 2 is controlling the injection of the piezo injector 3. For this reason, it is desirable to obtain the switching loss using some alternative means. The inventor repeated various experiments in order to obtain this switching loss. It has been found that this switching loss depends on the energy released from the charging capacitor 7 and the capacitance value Cinj of the piezo element 4. Furthermore, it has been found that the circuit loss PL has a positive correlation with the above-described sum ΣIpeak of the peak currents of the piezo driving current applied multiple times in the form of pulses.

  That is, according to the inventors, when the fuel injection control device 2 increases the charge amount of the piezo element 4, the discharge charge Qdc of the charging capacitor 7 relatively increases. As a result, the inventors observed the result that the circuit loss PL increases because the peak current value Ipeak flowing through the piezo element 4 increases as a whole. For this reason, it can be estimated that there is a correlation between the peak current value Ipeak and the circuit loss PL. On the other hand, it has also been observed that when the capacitance value Cinj of the piezo element 4 increases, the peak current value Ipeak increases as a whole, so that the circuit loss PL increases. Also from this tendency, it can be estimated that there is a correlation between the peak current value Ipeak and the circuit loss PL.

FIG. 3 shows the correlation between the sum ΣIpeak of peak currents flowing through the piezo element 4 and the circuit loss PL. As shown in FIG. 3, it has been found that there is a positive correlation between the total peak current ΣIpeak of the piezo drive current and the circuit loss PL. FIG. 3 shows the characteristics of “thin solid line”, “thick solid line”, and “dashed line”. The “thin solid line” indicates the characteristics when the capacitance value Cinj of the piezo element 4 is a constant value (= several μF) that is relatively lower than the standard value. A “thick solid line” indicates characteristics when the capacitance value Cinj of the piezo element 4 is a standard value (= several μF). A “dashed line” indicates characteristics when the capacitance value Cinj of the piezo element 4 is a constant value (= several μF) that is relatively higher than the standard value. A “dotted line” indicates a linear approximation line along these characteristics. As shown in FIG. 3, it can be seen that there is a positive correlation between the circuit loss PL and the sum of the peak currents ΣIpeak. Considering such a relationship, the circuit loss of the fuel injection control device 2 is determined by determining α and β satisfying the approximate expression of the following equation (1) for each piezo injector 3 and its fuel injection control device 2. PLes can be estimated from the following equation (1).
PLes = α × ΣIpeak + β (1)

  That is, the circuit loss PLes can be estimated from the peak current value Ipeak obtained during actual driving without depending on the capacitance value Cinj of the piezo element 4. Further, the drive current of the piezo element 4 can be feedback controlled based on the estimated value of the circuit loss PLes of the equation (1).

  Here, a positive correlation between the sum ΣIpeak of peak currents and the circuit loss PL is shown. “Sum of peak currents ΣIpeak” is easy to imagine a large number of peak currents, but considering this in reverse, there is also a positive correlation between one or more peak current values Ipeak and circuit loss PLes. It can be said that it plays.

<Outline of Correction Processing of Correction Calculator 26>
FIG. 4 schematically shows the content of the correction processing of the correction calculator 26 shown in FIG. As shown in FIG. 4, the storage unit 6a includes an ejection information register 6aa. Here, the injection information register 6aa includes the charge current command values Qdc_N and Qdc_N + 1 and the pre-correction charge charge command value Qdc_N for post-injection together with the sum of the peak currents ΣIpeak_N and ΣIpeak_N + 1 of the past injection information. An area is secured so that +2 and ΣIpeak_N + 2 for calculating a correction value can be stored. The sum ΣIpeak_N of peak currents indicates the sum obtained by integrating the peak current values Ipeak during a period in which each charge on pulse P generated a predetermined number of times during the Nth charge control of the injection command is ON.
The sum ΣIpeak_N + 1 of the peak currents represents the sum total of the peak current values Ipeak during a period in which each charge on pulse P is on (ON) during the N + 1th charge control of the injection command. ing. The charge charge command value Qdc_N indicates a charge charge command value used for control corresponding to the Nth injection command signal in the order of the injection commands. The charge charge command value Qdc_N + 1 indicates the charge charge command value Qdc used for control corresponding to the N + 1th injection command signal in the order of the injection command. These injection information registers 6aa are based on a minimum storage capacity. The storage capacity of the ejection information register 6aa may be secured so as to be able to store an information amount exceeding this.

<Description of operation and operation>
The operation and operation of the above configuration will be described below with reference to FIGS. FIG. 5 shows a flowchart of the update process, and FIG. 6 shows a method for estimating the sum ΣIpeak_N + 2 of the peak current from the past injection information and the charge charge command value Qdc_N + 2. FIG. 7 shows a timing chart at the time of injection.

  Here, the injection information register 6aa stores the sum of peak currents ΣIpeak_N, ΣIpeak_N + 1, and the charge charge command given from the microcomputer 5 in response to the Nth and N + 1th injection command signals. Description will be made on the assumption that the values Qdc_N and Qdc_N + 1 are stored. See S0 in FIG.

When the microcomputer 5 outputs the N + 2th charge charge command value Qdc_N + 2 to the charge control circuit 17, the charge information command value Qdc_N + 2 is stored in the injection information register 6aa of the storage unit 6a. When the N + 2th charge charge command value Qdc_N + 2 is input from the microcomputer 5 to the injection information register 6aa (YES in S1 of FIG. 5), the correction calculator 26 performs a correction calculation using the charge charge command value Qdc_N + 2. Then, the charge charge correction command value Qdc_N + 2_adjust is calculated (S2 in FIG. 5). In S2 of FIG. 5, in particular, the correction calculator 26 reads the sum of peak currents ΣIpeakN + 1, ΣIpeakN and the charge charge command values Qdc_N, Qdc_N + 1 from the injection information register 6aa, and obtains the charge charge correction command value Qdc_N + 2_adjust. The correction is calculated and output to the threshold setting circuit 24. At this time, the correction calculator 26 corrects and calculates the charge charge correction command value Qdc_N + 2_adjust from the charge charge command value Qdc_N + 2 using the following equation (2).
Qdc_N + 2_adjust = Qdc_N + 2 × A (N + 1, N) + B (N + 1, N) (2)

In this equation (2), A (N + 1, N) and B (N + 1, N) are linear approximations based on past injection information ΣIpeak_N + 1, ΣIpeak_N, Qdc_N + 1, and Qdc_N. It is a parameter. These A (N + 1, N) and B (N + 1, N) indicate the gain term and offset term of the variation correction coefficient of the circuit loss PL, respectively. These A (N + 1, N) and B (N + 1, N) can be calculated, for example, by developing the following relational expression. The target N + 2 charging energy target value En_N + 2 of the injection command of the piezo injector 3 corresponds to a value obtained by subtracting the circuit loss PL from the energy released from the charging capacitor 7 (= Vdc × Qdc_N + 2). Then, the circuit loss PL before variation correction can be expanded by the following equation (3), and the circuit loss PL after variation correction (PL → PLes_N + 2) can be expanded as shown by the following equation (4). .
Vdc × Qdc_N + 2−PL = En_N + 2 (3)
Vdc x Qdc_N + 2_adjust-PLes_N + 2 = En_N + 2 (4)
Vdc and PL in equations (3) and (4) are constant values.
When this equation (4) is expanded by α and β using equation (1), equation (5) is obtained.
Vdc × Qdc_N + 2_adjust− (α × ΣIpeak_N + 2 + β) = En_N + 2 (5)

Based on the sum of past peak currents ΣIpeak_N + 1, ΣIpeakN and charge charge command values Qdc_N, Qdc_N + 1, the peak current when the order of injection commands is controlled by the charge charge command value Qdc_N + 2 for the (N + 2) th time The correlation for estimating the sum ΣIpeak_N + 2 can be expressed by linear approximation as shown in FIG. Based on this correlation, the following relationship (6) is established.
ΣIpeak_N + 2
= (ΣIpeak_N + 1−ΣIpeak_N) / (Qdc_N + 1−Qdc_N) × Qdc_N + 2
+ (ΣIpeak_N × Qdc_N + 1−ΣIpeak_N + 1 × Qdc_N) / (Qdc_N + 1−Qdc_N) (6)

In the first term on the right side of equation (6), if the gradient of the charge charge command value Qdc_N + 2 is Y (N + 1, N) and the intercept of the second term on the right side is δ (N + 1, N), The expression 6) can be expressed as the expression (7).
ΣIpeak_N + 2 = Y (N + 1, N) × Qdc_N + 2 + δ (N + 1, N) (7)

Therefore, when this equation (7) is substituted into ΣIpeak_N + 2 in the equation (5), it can be expressed as the following equation (8).
Vdc × Qdc_N + 2_adjust
-Α × {Y (N + 1, N) × Qdc_N + 2 + δ (N + 1, N)} + β = En_N + 2 (8)
If Qdc_N + 2_adjust is obtained from this equation (8), it can be developed as equation (9).
Qdc_N + 2_adjust
= [En_N + 2 + {α × δ (N + 1, N) + β}] / {Vdc−α × Y (N + 1, N)} (9)
When the charge charge correction command value Qdc_N + 2_adjust is obtained from the charge charge command value Qdc_N + 2 using this equation (9), it can be developed as shown in equation (10).
Qdc_N + 2_adjust
= Qdc_N + 2 × Vdc / {Vdc−α × Y (N + 1, N)}
+ [{Α × δ (N + 1, N) + β} −PL) / {Vdc−α × Y (N + 1, N)}] (10)

The slope of Qdc_N + 2 in the first term on the right side of equation (10) is A (N + 1, N), and the intercept of the second term on the right side in equation (10) is obtained as B (N + 1, N). (3) can be obtained. That is, the charge charge correction command value Qdc_N + 2_adjust can be obtained with respect to the charge charge command value Qdc_N + 2, thereby correcting variations in the circuit loss PLes_N + 2.
Although the first-order approximation formula is used here, it can also be expressed using a variety of approximation formulas such as polynomial approximation or logarithmic approximation. At this time, from the measurement data indicating the correlation of Qdc-ΣIpeak obtained in advance in the development design stage of the fuel injection control device 2, an approximate expression (for example, correlation coefficient> 0.9) that satisfies the required estimation accuracy is obtained. It is recommended to select an approximate expression that satisfies this condition.

When the correction calculator 26 calculates the charge charge correction command value Qdc_N + 2_adjust from the charge charge command value Qdc_N + 2 in S2 of FIG. 5 and sets it in the threshold setting circuit 24, the threshold setting is performed as shown in S3 of FIG. The circuit 24 sets a charge charge threshold value Qth_a for one charge in the charge on pulse P, and outputs a threshold voltage corresponding to the charge charge threshold value Qth_a to the comparison unit 25. At this time, the threshold setting circuit 24 sets the charge charge threshold Qth_a based on, for example, the equation (11).
Qth_a = Qth_a_base × Qdc_N + 2_adjust / Qdc_base (11)

  Here, Qdc_base indicates the standard value of the charge charge command value Qdc, and Qdc_a_base indicates the standard threshold value of the charge charge of the a-th charge on-pulse P. In other words, the standard threshold value Qth_a_base of the charge for charging is determined for each charge on pulse P, and the threshold setting circuit 24, for example, the discharge charge ratio (= Qth_a_base / Qdc_base) is multiplied by the charge charge correction command value Qdc_N + 2_adjust to calculate a charge charge threshold value Qth_a, and a threshold voltage corresponding to the charge charge threshold value Qth_a is output to the comparison unit 25. See timing t0 in FIG.

  The charge control circuit 17 corrects the variation in the circuit loss PL by the correction calculator 26, and the threshold setting circuit 24 waits for the injection command signal in a state where the threshold is set in the comparison unit 25. The microcomputer 5 outputs the active level “H” of the injection command signal. Then, when the control circuit 6 inputs the active level “H” of the injection command signal, it becomes YES in S4 of FIG. 5 and starts the charging control of the piezo injector 3, and the charging control circuit 17 activates the active level of the charging period command signal. “H” is output. See timing t1 in FIG. Then, the charge on control circuit 20 outputs the charge on signal of the charge on pulse P as described above. See timing t2 in FIG.

  When the charge measurement circuit 23 receives the charge measurement permission signal EN accompanying the charge on signal of the charge on pulse P, the charge measurement circuit 23 integrates the current flowing through the current detection resistor 15 through the piezo element 4 of the piezo injector 3 to thereby measure the charge Qinj_a (“a "" Starts measurement of the number of charging on pulses P).

  The comparison unit 25 compares the voltage corresponding to the measurement charge Qinj_a obtained by integrating the current flowing through the current detection resistor 15 with the threshold voltage obtained by the above equation (11), but corresponds to one measurement charge Qinj_a. When the voltage becomes equal to or higher than the threshold voltage corresponding to the charge charge threshold value Qth_a of the threshold setting circuit 24, the logic gate 21 outputs “L” to the logic gate 21, and the logic gate 21 controls the charge switch 8 to be turned off through the voltage buffer 12. See timing t3 in FIG.

The charging current of the piezo injector 3 has a peak value at timing t3 when the charging switch 8 is turned off. At this timing t3, the peak calculator 27 charges the falling signal of the charging signal output from the logic gate 21. Since it is input as an OFF signal, it is determined YES in S6 of FIG. 5, and in S7 of FIG. 5, the peak current value Ipeak_a flowing through the current detection resistor 15 is acquired, and the peak current value Ipeak_a is expressed by the following equation (12). And the summation ΣIpeak_N + 2 of the peak current is updated in the injection information register 6aa of the storage unit 6a. See S8 in FIG.
ΣIpeak_N + 2 = ΣIpeak_N + 2 + Ipeak_a (12)

  Thereafter, the steps S6 to S9 in FIG. 5 are repeated until the number “a” of the charge on pulses P reaches a predetermined number, and the final sum of peak currents ΣIpeak_N + 2 is calculated.

Thereafter, when the number “a” of charge on pulses P reaches a predetermined number, the charge period is completed (YES in S9 of FIG. 5), and the stored value of the injection information register 6aa is updated by a queue method as described below (FIG. 5). S10).
ΣIpeak_N: = ΣIpeak_N + 1 (12)
ΣIpeak_N + 1: = ΣIpeak_N + 2 (13)
Qdc_N: = Qdc_N + 1 (14)
Qdc_ (N + 1): = Qdc_N + 2_adjust (15)

  That is, the sum ΣIpeak_N + 1 of the (N + 1) th peak current is updated as the sum ΣIpeak_N of the Nth peak current as shown in the equation (12), and the (N + 2) th as shown in the equation (13). The sum ΣIpeak_N + 2 of the peak currents is sequentially updated by the cue method as the sum ΣIpeak_N + 1 of the (N + 1) th peak current. Further, as shown in the equation (14), the (N + 1) th charge charge command value Qdc_N + 1 is updated as the Nth charge charge command value Qdc_N, and as shown in the equation (15), the (N + 2) th charge charge is obtained. The command value Qdc_N + 2 is sequentially updated by the cue method as the (N + 1) th charge charge command value Qdc_N + 1. Thus, the update unit 28 performs the update process, so that preparations for correcting the charge charge command value Qdc input corresponding to the next injection command can be made.

  Then, the updating unit 28 clears the value of ΣIpeak_N + 2 in the injection information register 6aa in S11 of FIG. Further, since the charge off signal of the charge on pulse P is given to the charge measurement circuit 23 as a measurement charge clear signal, the charge measurement circuit 23 clears the measurement charge Qinj_a. See timings t4, t7, and t10 in FIG.

  The charge control circuit 17 returns to S1 in FIG. 5 and repeats the process, but again calculates Qdc_N + 2_adjust in the process of S1 to S3 by inputting the charge charge command value Qdc_N + 2 corresponding to the next injection command. Thus, in preparation for the next injection command, the charged charge threshold value Qth_a corrected for one charge on pulse P is set in advance. Then, when the active level “H” of the next injection command is input in S4, the charging control circuit 17 performs charging control by the processing of S5 to S11 in a state where the variation of the circuit loss PL is corrected, and obtains ΣIpeak_N + 2. Update and prepare for correction control in the next injection command.

  In this way, correction control logic S1 to S3 for variation in circuit loss PL and update logic S6 to S11 for variation coefficient for variation in circuit loss PL can be corrected in combination.

  As described above, according to the present embodiment, the sum ΣIpeak_N of the peak currents of the currents flowing through the piezo element 4 when the charge control circuit 17 performs charge control on the piezo element 4 in response to the charge on pulse P, The charge charge command value Qdc_N + 2 input during post-injection is corrected using ΣIpeak_N + 1, charge charge command values Qdc_N, and Qdc_N + 1 as past results. As a result, the charge charge command value Qdc_N + 2 can be corrected in consideration of the influence of the circuit loss PL, the piezo injector 3 can be controlled to be opened with high accuracy, and the injection accuracy of the fuel injection control device 2 can be improved.

(Other embodiments)
The present invention is not limited to the above-described embodiments, can be implemented with various modifications, and can be applied to various embodiments without departing from the gist thereof. For example, the following modifications or expansions are possible. The plurality of embodiments described above can be combined.

  Using the sum of peak currents ΣIpeak_N and ΣIpeak_N + 1 as past results, the charge charge correction command value Qdc_N + 2_adjust is calculated from the next charge charge command value Qdc_N + 2 based on the calculation formula shown in the above embodiment. Although shown, it is not limited to this.

  In particular, among the past results of a plurality of peak current values Ipeak obtained corresponding to a plurality of charge on pulses P, the subsequent charge charge command value Qdc_N is based on the past results of at least one or more peak current values Ipeak. Correct +2.

  As a past result, for example, after the active level “H” of the charging period command signal is input, the peak current value Ipeak corresponding to the first charging on pulse P and before the charging period command signal becomes the non-active level “L”. The peak current value Ipeak corresponding to the last charge on pulse P may be used, and this past result can be used to correct the charge charge command value Qdc at the time of post-injection. At this time, it is desirable to correct the charge charge command value Qdc at the time of post-injection based on the difference between the first and last peak current values Ipeak among the peak current values Ipeak corresponding to all the charge on pulses P.

  Further, for example, among all the charge on pulses P in one charge period command signal, the peak current value Ipeak at which the energization current becomes the maximum value and the minimum value is used as a past result, and charging at the time of subsequent injection is performed. The present invention can also be applied to a form in which the charge command value Qdc is corrected. At this time, it is desirable to correct the charge charge command value Qdc at the time of subsequent injection by the difference obtained by subtracting the minimum value from the maximum value of the peak current value Ipeak among the peak current values Ipeak corresponding to all the charge on pulses P.

  Further, only one or two peak current values Ipeak are acquired in a plurality of charge on pulses P in one charge period command signal “L” → “H” → “L”, and this is stored in the past. As a result, the charge charge command value Qdc may be corrected.

  In such a case, since the number of detected peak currents can be reduced as compared with the above-described embodiment, the number of processes can be reduced. For example, when configured using hardware, the circuit scale can be reduced as much as possible.

  In the above-described embodiment, the correction is performed based on a predetermined arithmetic expression, but the present invention is not limited to this. For example, the correction calculator 26 uses any arithmetic expression related to nonlinear conversion in addition to linear conversion such as four arithmetic operations of addition, subtraction, multiplication, and division for the input charge charge command value Qdc_N + 2 and the peak current value Ipeak described above. Then, the charge charge correction command value Qdc_N + 2_adjust may be calculated.

  Instead of the microcomputer 5 and the control circuit 6, various control devices may be used. Means and / or functions provided by the control device can be provided by software recorded in a substantial memory device and a computer that executes the software, software, hardware, or a combination thereof. For example, when the control device is provided by an electronic circuit that is hardware, the control device can be configured by a digital circuit including one or a plurality of logic circuits, or an analog circuit. Further, for example, when the control device executes various controls by software, a program is stored in the storage unit, and a method corresponding to the program is executed when the control subject executes the program.

  The reference numerals in parentheses described in the claims indicate the correspondence with the specific means described in the embodiment described above as one aspect of the present invention, and limit the technical scope of the present invention. It is not a thing. An aspect in which a part of the above-described embodiment is omitted as long as the problem can be solved can be regarded as the embodiment. In addition, any conceivable aspect can be regarded as an embodiment as long as it does not depart from the essence of the invention specified by the words described in the claims.

  Moreover, although this invention was described based on embodiment mentioned above, it understands that this invention is not limited to the said embodiment and structure. The present invention includes various modifications and modifications within an equivalent range. In addition, various combinations and forms, as well as other combinations and forms including one element, more or less, are within the scope and spirit of the present disclosure.

  In the drawing, 2 is an electronic control device, 3 is a piezo injector (injector), 4 is a piezo element (capacitive actuator), 6a is a storage unit, 17 is a charge control circuit (charge control unit), and 26 is a correction calculator. Show.

Claims (3)

A fuel injection control device for controlling fuel injection by an injector (3) using a capacitive actuator (4),
A charge control unit (17) that generates a charge on-pulse at the time of fuel injection based on the input charge charge command value and discharges the charge accumulated in the charge accumulation unit to the capacitive actuator, thereby charging the capacitive actuator. )
The charge controller is
A correction calculator (26) for correcting a charge charge command value input at the time of post-injection based on a past result of charging the capacitive actuator with a charge,
The correction calculator uses the peak current value of the current flowing through the capacitive actuator as the past result when the charge control unit performs charge control on the capacitive actuator in response to the charge on pulse. A fuel injection control device that corrects a charge charge command value input during post-injection.   When the charge control unit (17) generates the charge on-pulse a plurality of times and controls the charge to the capacitive actuator, the correction calculator (26) receives each peak current value flowing through the capacitive actuator. The fuel injection control device according to claim 1, wherein the charge charge command value input at the time of the post-injection is corrected using the sum of the values as the past result.   When the charge control unit (17) generates the charge on-pulse a plurality of times and controls the charge to the capacitive actuator, the correction calculator (26) has at least two or more flowing through the capacitive actuator. The fuel injection control device according to claim 1, wherein a charge charge command value input at the time of the post-injection is corrected using a result calculated based on a predetermined arithmetic expression using a peak current value as the past result.

Images