JP2012172567A - Automatic start control device for internal combustion engine - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an automatic start control device for an internal combustion engine, predicting a minimum value of voltage in the next automatic start with high accuracy and rationalizing determination of permission with respect to an automatic stop request.SOLUTION: The automatic start control device for the internal combustion engine includes: a battery characteristic learning means which calculates the internal resistance Rin (characteristic value) and the maximum amount of discharge current ΔImax (characteristic value) of a battery based on battery voltage detected in the automatic start, and which stores the calculated characteristic values as learning values; and an automatic stop permission determination means which predicts the minimum value of voltage of the battery in the next automatic start based on the learning values when automatic stop is required, and which determines whether the request is permitted based on the predicted value. If the automatic start is performed within a predetermined time Tth after the automatic stop is performed (S54: YES), learning by the battery characteristic learning means in the automatic start is prohibited (S56).

Description

  The present invention relates to an automatic start control device that performs control to automatically start an internal combustion engine.

  In addition to supplying power to a starter that applies initial rotation to the output shaft of an engine (internal combustion engine), a battery mounted on a vehicle or the like can also supply power to various electronic control devices, navigation devices, audio devices, and the like. Is going. The battery voltage should be equal to or higher than the minimum voltage necessary for ensuring the reliability of operation of these devices.

  For this reason, even if power can be supplied to the starter when the engine is automatically started, the reliability of the operation of the control device or the like may not be ensured if the battery voltage drop caused by the starter driving is large. There is. Therefore, when the engine automatic stop is requested while the vehicle is running, it is desirable that the minimum value of the battery voltage at the next automatic start is predicted, and that the automatic stop is not permitted if the predicted minimum value is less than or equal to the threshold value. By the way, if the predicted minimum value is less than or equal to the threshold value during idle stop, the engine is automatically started.

  Here, the minimum voltage of the battery at the time of automatic start is expressed by the equation “ΔImax · R + V” according to the current internal resistance R of the battery, the maximum discharge current amount ΔImax of the battery at the start, and the current voltage V of the battery. Can be predicted.

  The maximum discharge current amount is a discharge amount (inrush current amount) when the current flowing through the starter rapidly increases immediately after energization of the starter is started. Here, since the maximum discharge current amount ΔImax changes according to the degree of deterioration over time such as the discharge capability of the battery, using this as a predetermined fixed value causes a decrease in the accuracy of the prediction.

  On the other hand, it is also difficult to detect the maximum discharge current amount ΔImax at the start in order to obtain an accurate value as the maximum discharge current amount ΔImax. That is, the maximum discharge current amount ΔImax takes a very large value that can hardly occur as the discharge current amount of the battery other than at the time of starting. And since the current sensor which can detect this amount of discharge current is very expensive, it is not preferable from the viewpoint of cost performance to provide such an expensive sensor only for starting.

  In view of these points, Patent Documents 1 and 2 sample a plurality of detected values of battery voltage and battery current during a period in which the discharge current amount of the battery decreases after reaching the maximum discharge current amount ΔImax ( Based on these detected values, it is described that characteristic values such as the internal resistance R and the maximum discharge current amount ΔImax are calculated and learned based on these detected values. According to this, when the automatic stop is requested, the voltage minimum value at the next automatic start can be predicted based on the learned value, and the permission determination for the request can be made appropriate.

JP 2009-114939 A JP 2010-24906 A

  However, if the automatic start is performed immediately after the automatic stop, the characteristic value is erroneously learned as follows, and there arises a problem that the permission determination cannot be properly performed. Found. Hereinafter, the cause of the erroneous learning will be described.

  FIG. 7 is a graph showing test results conducted by the inventor. The horizontal axis represents the elapsed time from the engine stop point, and the vertical axis represents the change in battery voltage. The test result indicated by reference sign LA in the figure is a case where the engine is stopped while the battery is being charged. When the battery is charged, the battery voltage is apparently high due to the influence of the charge polarization. Therefore, when the battery is no longer charged as the engine stops, the battery voltage gradually decreases as indicated by symbol LA. At this time, the voltage decrease rate is high for a few seconds immediately after the engine is stopped (immediately after the end of charging), and then the voltage decrease rate is gradually decreased.

  Here, in the inventions described in Patent Documents 1 and 2, during the period in which the discharge current amount of the battery decreases after reaching the maximum discharge current amount ΔImax, a plurality of detected values of the battery voltage and the battery current are sampled. As described above, the characteristic value is calculated and learned based on the detected value.

  However, if the sampling is performed during the period when the battery voltage decrease rate is high as described above, the degree of influence of the charge polarization received by the detection value at the initial stage of the sampling (see S1 in FIG. 4) and the end of sampling are detected. The degree of influence of charge polarization on the value (see symbol S2 in FIG. 4) is greatly different. Then, the amount by which the initial detection value S1 is apparently higher is larger than the amount by which the final detection value S2 is apparently higher. Therefore, the characteristic values calculated based on these sampling detection values greatly deviate from the actual values. End up mis-learning.

  For example, when the slope of the voltage-current characteristic (corresponding to the internal resistance) is calculated as the characteristic value based on the initial detection value S1 and the final detection value S2, the initial detection value S1 is increased due to the influence of polarization. As a result, the internal resistance is calculated to be smaller than the actual value.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to accurately predict the minimum voltage value at the next automatic start and optimize the permission determination for the automatic stop request. It is to provide an automatic start control device for an engine.

  Hereinafter, means for solving the above-described problems and the operation and effects thereof will be described.

  According to the first aspect of the present invention, a starter that applies an initial rotation to an output shaft of an internal combustion engine to start the battery, a battery that supplies power to the starter, and an idle stop system that automatically stops and automatically starts the internal combustion engine. It is assumed that it is applied to the engine system provided.

  Then, the battery characteristic learning means for calculating the characteristic value of the battery based on the battery voltage detected at the time of the automatic start and storing the calculated characteristic value as a learning value, and when the automatic stop is requested next time Automatic stop permission determining means for predicting a minimum battery voltage at the time of automatic start based on the learned value and determining whether to permit the request based on the predicted value, and performing the automatic stop When the automatic start is performed within a predetermined time after the start, learning by the battery characteristic learning means at the time of the automatic start is prohibited.

  According to this, learning of the characteristic value based on the battery voltage detected when the influence of the battery polarization is large is prohibited, so that erroneous learning due to the battery polarization can be avoided, and the determination by the automatic stop permission determination means is made appropriate. Can be planned.

  According to a second aspect of the present invention, a starter that applies an initial rotation to an output shaft of an internal combustion engine to start the battery, a battery that supplies power to the starter, and an idle stop system that automatically stops and automatically starts the internal combustion engine. It is assumed that it is applied to the engine system provided.

  Then, the battery characteristic learning means for calculating the characteristic value of the battery based on the battery voltage detected at the time of the automatic start and storing the calculated characteristic value as a learning value, and when the automatic stop is requested next time Automatic stop permission determining means for predicting a minimum battery voltage at the time of automatic start based on the learned value and determining whether to permit the request based on the predicted value, and performing the automatic stop The battery voltage minimum value predicted based on the learned value at the time of the automatic start, or learned by correcting the characteristic value calculated at the time of the automatic start It is characterized by correcting.

  According to this, for the characteristic value (apparent characteristic value) based on the battery voltage detected when the influence of the battery polarization is large, the influence characteristic of the battery polarization is taken into account to correct the apparent characteristic value, or learn, The battery voltage minimum value predicted based on the apparent characteristic value is corrected in consideration of the effect of battery polarization. Therefore, it is possible to determine whether or not to permit the automatic stop request in consideration of the influence of the battery polarization, so that the determination by the automatic stop permission determination unit can be optimized.

  According to a third aspect of the present invention, when the apparent battery voltage due to the polarization of the battery changes with the recovery of polarization from the time when the automatic stop is performed, the change per unit time of the battery voltage The time required for the amount to become less than a predetermined value is the predetermined time.

  According to this, when the apparent battery voltage due to the polarization of the battery changes with the recovery of polarization after the automatic stop, the change amount (change speed) per unit time of the battery voltage becomes less than a predetermined value. The time required until the time until the battery voltage change rate becomes less than a predetermined value can be set as the predetermined time based on the test result. Therefore, it can be reliably avoided that the determination by the automatic stop permission determination means is improperly performed due to the influence of the battery polarization.

  According to a fourth aspect of the present invention, an integrated value of the current flowing through the battery during a predetermined period is calculated, and the predetermined time is variably set according to the integrated value.

  The inventor has tested how the battery voltage change rate varies when the battery voltage changes with the recovery of polarization when the integrated value (integrated current) of the battery current is different. As a result, it was found that the battery voltage change rate became faster as the integrated current increased (see FIG. 7). In the above invention in view of this point, since the predetermined time is variably set according to the integrated current, it is possible to appropriately determine whether to prohibit learning or whether to perform correction. In particular, when the “time required until the amount of change per unit time of the battery voltage becomes less than a predetermined value” is set as the predetermined time as in claim 3, such setting can be realized with high accuracy. Specific examples of the “predetermined period” include a period from the previous automatic start to the current automatic stop, a period from when the ignition switch is turned on to the current automatic stop, and the like.

  According to a fifth aspect of the present invention, when the automatic start is performed within a second predetermined time after the power supply from the battery to a predetermined electric load is cut off, the battery characteristic learning at the time of the automatic start is performed. It is characterized by prohibiting learning by means.

  By the way, in the inventions according to claims 1 to 4 described above, “when charging is stopped during battery charging, the influence of polarization is large within a predetermined time from the charging stop, so there is a problem of erroneous learning. We pay attention to the point that it occurs. On the other hand, in the invention according to claim 5, “if the discharge is stopped during the battery discharge, the influence of the polarization is large if it is within the second predetermined time from the stop of the discharge. ”.

  That is, for example, immediately after the headlight is turned off during the idling stop or the air conditioner is turned off (immediately after the end of the discharge), the battery voltage gradually increases as indicated by the symbol LB in FIG. At this time, the voltage rise rate is high for a few seconds immediately after the end of the discharge, and then the voltage rise rate gradually decreases. Therefore, for example, in the case where the slope of the voltage-current characteristic (corresponding to the internal resistance) is calculated as the characteristic value as described above, the initial detection value S1 (see FIG. 4) is lowered due to the influence of polarization. The resistance is calculated to be larger than the actual value.

  In view of this point, in the invention according to claim 5, within a second predetermined time after the power supply to a predetermined electric load (for example, a headlight or an air conditioner) is cut off (when the influence of battery polarization is large). Since learning of the characteristic value based on the detected battery voltage is prohibited, mislearning due to battery polarization can be avoided, and the determination by the automatic stop permission determining means can be optimized.

  In the invention according to claim 6, when the automatic start is performed within the second predetermined time after the power supply from the battery to the predetermined electric load is cut off, the characteristic value calculated at the time of the automatic start is calculated. Or the battery voltage minimum value predicted based on the learning value at the time of automatic start is corrected.

  According to the above invention, the characteristics based on the battery voltage detected within the second predetermined time (when the influence of the battery polarization is large) after the power supply to the predetermined electric load (for example, headlight, air conditioner, etc.) is cut off. The value (apparent characteristic value) is learned by correcting the apparent characteristic value in consideration of the influence of the battery polarization, or the battery voltage minimum value predicted based on the apparent characteristic value is considered in consideration of the influence of the battery polarization. To correct. Therefore, it is possible to determine whether or not to permit the automatic stop request in consideration of the influence of the battery polarization, so that the determination by the automatic stop permission determination unit can be optimized.

The invention according to claim 7 is characterized in that the second predetermined time is set to a time shorter than the predetermined time.
Here, as illustrated in FIG. 7, the rate of increase of the battery voltage immediately after stopping the discharge (see symbol LB) is slower than the rate of decrease of the battery voltage immediately after stopping charging (see symbol LA). That is, the influence of polarization is less immediately after stopping the discharge than when immediately after stopping the charge. In the above invention in view of this point, the second predetermined time immediately after the stop of discharge, which is the subject of learning prohibition (or correction), is set to be shorter than the predetermined time immediately after the stop of charging. It is possible to promote the optimization of the determination as to whether or not to make it.

The figure which shows the system configuration | structure concerning 1st Embodiment of this invention. The figure which shows transition of the battery discharge current at the time of starting. The figure explaining the estimation principle of the internal resistance at the time of the maximum discharge current concerning the said embodiment. The figure explaining the presumed principle of the maximum discharge current concerning the embodiment. The flowchart which shows the procedure of the estimation process of the maximum discharge current concerning the embodiment. The flowchart which shows the procedure of the prohibition process of the engine automatic stop start process concerning the embodiment. It is a figure which shows the change of the battery voltage immediately after an engine stop (just after completion | finish of battery charging / discharging), Comprising: The figure showing the influence of polarization. The figure which shows the test result which shows the difference | error (prediction error) of the minimum voltage Vmin calculated based on Rin and (DELTA) Imax which received the influence of polarization, and the minimum voltage Vmin actually measured and tested. 6 is a flowchart showing a procedure for determining whether to permit learning of Rin and ΔImax based on a time threshold value (predetermined time Tth) in the embodiment. The flowchart which shows the procedure of the prohibition process of the engine automatic stop start process concerning 3rd Embodiment of this invention.

  Embodiments of an automatic start control device for an internal combustion engine according to the present invention will be described below with reference to the drawings. In the following embodiments, parts that are the same or equivalent to each other are denoted by the same reference numerals in the drawings, and the description of the same reference numerals is used.

(First embodiment)
The automatic start control device according to the present embodiment controls an internal combustion engine mounted on a vehicle as a travel drive source.

  FIG. 1 shows the overall system configuration of the internal combustion engine according to the present embodiment.

  The internal combustion engine 10 is a port injection type gasoline engine. The internal combustion engine 10 is a power generation device for a vehicle, and its output shaft (crankshaft 12) is mechanically connected to drive wheels. On the other hand, the power generation apparatus 20 includes an alternator 22 as a generator and a regulator 24 as a control circuit that controls the output of the alternator 22. Here, the rotor of the alternator 22 is mechanically connected to the crankshaft 12 of the internal combustion engine 10 and is rotated by the rotational force of the crankshaft 12.

  A battery 30 as a lead storage battery is connected to the battery terminal TB of the power generation device 20. An electric load 44 is connected to the battery 30 via the switch 42 in parallel. Furthermore, a starter 40 that applies initial rotation to the crankshaft 12 of the internal combustion engine 10 is connected to the battery 30 as an electrical load. On the other hand, the power supply line between the battery terminal TB and the battery 30 and the ignition terminal TIG of the power generation apparatus 20 are connected via an ignition switch 46.

  The electronic control unit (ECU 50) as one of the electric loads of the battery 30 is mainly composed of a microcomputer and includes a storage device such as a constant storage holding device 51. Here, the constant memory holding device 51 is a memory device that always holds the memory regardless of the state of the start switch of the engine control system (for example, the ignition switch 46 that is the main power switch of the ECU 50). Specifically, for example, there is a backup RAM in which the power supply state is always maintained regardless of the state of the start switch, or a nonvolatile memory such as an EEPROM (registered trademark) that always holds a memory regardless of whether power is supplied.

  The ECU 50 controls the internal combustion engine 10 and the power generation device 20. In particular, in the ECU 50, the detection value of the current sensor 52 that detects the current discharged from the battery 30 and the current charged to the battery 30, the detection value of the temperature sensor 54 that detects the temperature of the battery 30, and further the battery 30. The voltage applied to the battery terminal TB of the power generation device 20 (the output voltage of the power generation device 20) is controlled based on the detection value of the voltage sensor 56 that detects the voltage of. Specifically, the ECU 50 outputs a command value (command voltage) of the output voltage to the command terminal TR of the power generation device 20. Thereby, the regulator 24 controls the output voltage to the command voltage. Further, the ECU 50 takes in a power generation state signal indicating the power generation capability of the power generation device 20 via the monitor terminal TF of the power generation device 20. Here, the power generation capacity is quantified by the ON / OFF time ratio of the switching element in the regulator 24 (specifically, the ratio of the ON time to the ON / OFF cycle: Duty).

  The control of the output voltage is performed so as to reduce the increase in the fuel consumption of the internal combustion engine 10 by the power generation by the power generation device 20 as much as possible while keeping the state of charge (SOC) of the battery 30 within an allowable range. Is called. Here, the SOC is a physical quantity obtained by quantifying the discharge capacity of the battery 30. Specifically, the ratio of the current charge amount to the full charge of the battery 30 is quantified. The SOC is usually quantified by “5 hour rate capacity”, “10 hour rate capacity” or the like. It is known that the open end voltage, which is the voltage when the terminal of the battery 30 is open, depends on the SOC. Specifically, the open end voltage increases as the SOC increases. Specifically, for example, the open circuit voltage when the SOC is “100%” is “12.8 V”, and the open circuit voltage when the SOC is “0%” is “11.8 V”.

  Further, the ECU 50 stops the idle rotation speed control of the internal combustion engine 10 and stops the internal combustion engine 10 automatically when the vehicle is stopped, so-called idle stop control (automatic stop processing), or the internal combustion engine 10 from the idle stop control. An automatic start process for automatically starting is performed. Here, the automatic start process is executed by performing combustion control after applying the initial rotation to the crankshaft 12 of the internal combustion engine 10 by starting the starter 40. However, when starting the starter 40, it is known that a large amount of discharge current flows from the battery 30 to the starter 40 in a very short time until the starter 40 starts rotating. At this time, the voltage of the battery 30 drops greatly. On the other hand, the ECU 50 using the battery 30 as a power supply means defines a minimum value of the power supply voltage that can maintain the reliability of the operation. For this reason, when the voltage of the battery 30 decreases excessively during the automatic start process of the internal combustion engine 10, the reliability of the operation of the ECU 50 may decrease. For this reason, it is desirable that the idle stop control be performed under the condition that the voltage drop of the battery 30 due to the automatic start process does not fall below the minimum value of the operation guarantee voltage of the ECU 50.

  Here, the minimum value of the voltage of the battery 30 that accompanies the automatic start process is considered to be a value that is taken when the amount of discharge current of the battery 30 is maximized. Therefore, in the present embodiment, when the internal combustion engine 10 is started by operating the ignition switch 46, information regarding the maximum discharge current amount of the battery 30 is calculated as a battery characteristic value, and the calculated characteristic value is stored as a learning value. The learning value is used when predicting the amount of decrease in the voltage of the battery 30 due to the automatic start process from the state of the battery 30 each time. Hereinafter, the learning principle of information related to the maximum discharge current amount will be described.

  FIG. 2A shows the behavior of the current of the battery 30 as the starter 40 is started. When the starter 40 and the battery 30 are electrically connected, the discharge current amount of the battery 30 rapidly increases and reaches the maximum value Imax at time t1. This maximum value Imax (inrush current to the starter 40) is determined by the resistance of the starter 40, the internal resistance of the battery 30, the wiring resistance between the starter 40 and the battery 30, and the like. When the starter 40 starts rotating, the current flowing through the starter 40 decreases. In other words, the discharge current of the battery 30 decreases. Here, the current sensor 52 does not set the maximum value Imax of the discharge current of the battery 30 when the starter 40 is activated within the detectable range. Therefore, in the present embodiment, the maximum discharge current amount of the battery 30 is estimated when the internal combustion engine 10 is started by operating the ignition switch 46.

  FIG. 2B shows changes in the current and voltage of the battery 30 when the starter 40 is activated. As shown in the figure, when the discharge current of the battery 30 increases with the starter 40 and reaches the maximum value Imax, the voltage of the battery 30 decreases to the minimum voltage Vbtm. At this time, since the influence of the polarization of the battery 30 is very small, the internal resistance of the battery 30 estimated from the voltage change with respect to the current change of the battery 30 is very close to the true value. However, since this period is very short, it is difficult to calculate the internal resistance by sampling current and voltage during this period.

  After the discharge current reaches the peak (maximum value Imax), the voltage of the battery 30 also increases as the discharge current decreases. Here, after the discharge current of the battery 30 reaches the maximum value Imax, the period during which the discharge current decreases is relatively long. Sampling is possible. For this reason, the internal resistance of the battery 30 can be calculated relatively easily during this period. On the other hand, it is generally considered that the internal resistance of the battery 30 during the period until the discharge current of the battery 30 reaches the maximum value Imax is different from the internal resistance after the maximum value Imax. Therefore, it is difficult to estimate the maximum value Imax based on the minimum voltage Vbtm of the battery 30 using the internal resistance of the battery 30 calculated based on the detected value of the current after the discharge current reaches the maximum value Imax. .

  Therefore, in this embodiment, the difference between the internal resistance Rc after reaching the maximum value Imax and the internal resistance Rin before reaching the maximum value Imax has a correlation with the open-circuit voltage difference ΔVo before and after starting the starter 40. Focus on the nature. FIG. 3 shows measurement data that supports this property. In FIG. 3, the vertical axis represents the difference between the internal resistance Rc (internal resistance before reaching the maximum value Imax) at the time of cranking and the internal resistance Rc (internal resistance after reaching the maximum value Imax) at the time of cranking. The horizontal axis indicates the difference ΔVo between the open end voltages before and after starting the starter. As shown in the figure, the larger the difference ΔVo, the smaller the difference of the internal resistance Rc relative to the internal resistance Rin. For this reason, it is possible to estimate the internal resistance Rin when the discharge current suddenly increases from the difference ΔVo and the internal resistance Rc.

  That is, first, based on the locus of the voltage of the battery 30 shown by a solid line in FIG. 4, the voltage Vt of the battery 30 before starting the starter 40 and the minimum voltage Vbtm of the battery 30 accompanying the starting of the starter 40 are detected. The voltage drop amount ΔV of the battery 30 due to the starter 40 being activated is calculated. On the other hand, the internal resistance Rc indicated by the one-dot chain line in the figure can be calculated from the behavior of the current and voltage after the discharge current reaches the maximum value. Then, by using this and the difference ΔVo between the open end voltages before and after the starter 40 is started, the internal resistance Rin at the time of sudden increase of the discharge current indicated by a two-dot chain line in the figure can be estimated. Then, the maximum discharge current amount ΔImax of the battery 30 due to the starter 40 being activated can be estimated and calculated from the internal resistance Rin and the drop amount ΔV.

  FIG. 5 shows a procedure for estimating the maximum discharge current amount ΔImax according to the present embodiment. This process is executed by the ECU 50 as a trigger when the ignition switch 46 is turned on.

  In this series of processing, first, in step S10, the voltage Vt and current It of the battery 30 which are detection values of the voltage sensor 56 and the current sensor 52 are acquired. Since it remains in step S10 until it is determined in the subsequent step S12 that the starter 40 has been activated, the voltage Vt and the current It may be average values of a plurality of detection values during this period. If it is determined in the subsequent step S12 that the starter 40 has been activated, the current and voltage of the battery 30 after the starter 40 is activated are detected a plurality of times in step S14. In the subsequent step S16, the voltage drop amount ΔV of the battery 30 is calculated by subtracting the minimum voltage Vbtm of the battery 30 detected in step S14 from the voltage Vt acquired in step S10.

  In the subsequent step S18, the internal resistance Rc of the battery 30 during cranking is calculated. Here, a plurality of sets of voltages and currents detected at the same time in step S14 are used. The plurality of sets are set of detection values during a period in which the discharge current of the battery 30 has risen to the lower limit value of the detectable range of the current sensor 52. This can be performed, for example, by detecting that the current of the battery 30 once decreases and then increases beyond the lower limit value of the detectable range. Using these plural sets, the internal resistance Rc can be estimated by a known regression analysis or the like. That is, a linear equation model with current as an explanatory variable and voltage as an objective variable is calculated by the method of least squares, and a coefficient of the linear equation model is set as an internal resistance Rc. In addition, it is good also considering the said several group as a predetermined number of groups selected from these instead of making it all the groups of applicable detection value.

  In the subsequent step S20, the difference ΔVo between the open-end voltage of the battery 30 before and after the starter 40 is started is calculated. Here, the open circuit voltage Voaf after the starter 40 is started can be an intercept of the primary model calculated by the regression analysis. On the other hand, the open-circuit voltage Vove before starting the starter 40 assumes that the internal resistance does not change before the starter 40 starts and when the discharge current suddenly increases, and approximates this with the internal resistance Rc. By using Vt, it can be simply estimated by the equation “Vobe = Vt−Rc · It”. Therefore, the open-circuit voltage difference ΔVo can be calculated as “Vobe−Voaf”.

  In the subsequent step S22, the internal resistance Rin at the time when the discharge current suddenly increases is estimated. Here, based on the correlation shown in FIG. 3, the internal resistance Rin is estimated by correcting the internal resistance Rc calculated in step S18 with the difference ΔVo calculated in step S20. . In the subsequent step S24, the voltage drop amount ΔV calculated in step S16 is divided by the internal resistance Rin estimated in step S22, whereby the maximum discharge current amount ΔImax of the battery 30 due to the starter 40 being started. Is calculated. In step S26 (battery characteristic learning means), the maximum discharge current amount ΔImax and the internal resistance Rin are stored (learned), and the detected values of the internal resistance Rc and the current and voltage are deleted. In addition, when the process of step S26 is completed, this series of processes is once complete | finished.

  FIG. 6 shows a processing procedure for prohibiting the idle stop control based on the maximum discharge current amount ΔImax. This process is repeatedly executed by the ECU 50 at a predetermined cycle, for example.

  In this series of processing, first, in step S30, it is determined whether or not an idle stop request is generated by satisfying a predetermined idle stop condition. Examples of the idle stop condition include that the SOC of the battery 30 is equal to or higher than a predetermined value, that the vehicle is stopped (vehicle speed is zero), or that the vehicle speed is less than the predetermined value and that the vehicle is decelerating.

  If it is determined that an idle stop request has occurred, the detected value of the voltage of the battery 30 is acquired in step S32. The detected value is preferably the latest value of the detected values of the voltage of the battery 30 so as to be a value corresponding to the current voltage value of the battery 30, but is arbitrary under the condition of the predetermined time. It may be a detected value.

  In subsequent step S34, a minimum voltage Vmin (battery voltage minimum value) associated with a voltage drop of the battery 30 when it is assumed that the internal combustion engine 10 has been restarted at the present time is predicted. Here, the maximum voltage drop amount is predicted based on the maximum discharge current amount ΔImax and the internal resistance Rin, which are learning values learned in the process of step S26 shown in FIG. That is, the minimum voltage Vmin is predicted by “Vmin = ΔImax · Rin + V”. In the subsequent step S36 (automatic stop permission determination means), it is determined whether or not the minimum voltage Vmin is equal to or lower than the threshold voltage Vth. This process determines whether or not the current state of the battery 30 is a state where the restart process of the internal combustion engine 10 can be appropriately performed. Here, the threshold voltage Vth is set based on the maximum value of the low voltages that cause a reduction in the reliability of the operation of the ECU 50.

  If it is determined that the voltage is equal to or lower than the threshold voltage Vth, the idle stop control is prohibited in step S38. That is, even if an idle stop request is generated when the vehicle is stopped or the like, the request is not permitted and the shift to the idle stop control is prohibited. On the other hand, if an affirmative determination is made in step S36 due to a voltage drop of the battery 30 or the like once the idling stop is permitted, the idling stop control is immediately stopped and the internal combustion engine 10 is turned off. Start immediately. From this point of view, it is desirable that the threshold voltage Vth be equal to or higher than the minimum voltage that can maintain the reliability of the operation of the ECU 50 when the internal combustion engine 10 is restarted. When the process of step S38 is completed or when a negative determination is made in steps S30 and S36, an idle stop is permitted in step S40, and this series of processes is temporarily ended.

  Thus, according to the present embodiment, when the internal combustion engine 10 is started by the starter 40 in the state of the battery 30 each time, the operation of the ECU 50 is predicted by predicting the lowest voltage Vmin that is assumed as the voltage of the battery 30. It is possible to perform the idle stop control until the voltage reaches the lower limit of the voltage that can maintain the reliability. For this reason, the fuel consumption of the internal combustion engine 10 can be reduced.

  Incidentally, it is assumed that the process of FIG. 6 is performed when the engine is running (for example, when the vehicle is running), but the following process is performed when the engine is stopped (for example, when the engine is idle stopped) separately from the process of FIG. It has been implemented. That is, during the idle stop, the same processes as in steps S32, S34, and S36 are repeatedly performed. If a negative determination is made in step S36, the idle stop is continued, and if a stroke determination is made in step S36, Automatically restart the engine.

  By the way, while the battery 30 is being charged, the open circuit voltage of the battery 30 is apparently high due to the polarization of the battery 30. Then, when the battery charging is terminated, the open circuit voltage gradually decreases as the polarization is gradually eliminated (see symbol LA in FIG. 7). At this time, immediately after the end of charging, the rate of decrease of the open circuit voltage is fast, the rate of decrease gradually decreases, and converges to the value of the open circuit voltage in the state without polarization.

  This means that if an idle stop is performed during battery charging and restarted immediately thereafter, the battery voltage detected by the voltage sensor 56 during the restart will appear to be higher due to the influence of polarization. means. Further, it means that the rate of decrease when the battery voltage, which is apparently high, decreases with the recovery of polarization, becomes faster as it is immediately after the restart.

  Therefore, if the idle stop is performed while the battery is being charged and restarted immediately thereafter, the internal resistance Rc calculated in step S18 in FIG. 5 is erroneously set to a value smaller than the actual value due to the influence of polarization. It will be calculated. That is, the amount that the detected value at the beginning of sampling (see symbol S1 in FIG. 4) is apparently higher than the amount that the detected value at the end of sampling (see symbol S2 in FIG. 4) is apparently higher. Many. Therefore, the inclination (internal resistance Rc) indicated by the one-dot chain line in FIG.

  Incidentally, the degree of influence of polarization is almost the same between the voltage Vt used for calculating the voltage drop amount ΔV in step S16 and the minimum voltage Vbtm. This is because the voltage drop time from Vt to Vbtm is extremely short. Therefore, since both the detected values of Vt and Vbtm are apparently higher by the same amount, the calculation result of the voltage drop amount ΔV substantially coincides with the actually generated drop amount.

  In short, the internal resistance Rc is calculated to be smaller than the actual value due to the influence of polarization, and the voltage drop amount ΔV is calculated to a value close to the actual value. Therefore, the maximum discharge current amount ΔImax (= ΔV / Rin) calculated in step S24 based on Rin and ΔV calculated from Rc is calculated to be larger than the actual value. Therefore, when an idle stop is performed during battery charging and restarted immediately thereafter, Rin and ΔImax calculated in steps S22 and S24 are different from actual values. Then, such different values are learned (mis-learned), the minimum voltage Vmin is predicted based on the learned values (S34), and whether or not the idle stop control is permitted is determined based on the predicted values (S36). There is a concern that an incorrect judgment will not be made.

  In response to these concerns, in the present embodiment, when restarting within a predetermined time Tth (see FIG. 7) from the time point when the idle stop is performed, the execution of the processing of FIG. It is avoided that the internal resistance Rin and the maximum discharge current amount ΔImax different from the above are learned.

  Regarding the setting method of the predetermined time Tth, as shown in FIG. 7, the elapsed time from the time when charging to the battery 30 is stopped and the battery voltage change are tested, and the time required for the voltage drop rate to become less than the predetermined value is determined. Set as a predetermined time Tth.

  Incidentally, symbols L1 to L4 in FIG. 7 indicate the results of performing the above test with different battery current integrated values (integrated currents). This result indicates that the voltage drop rate increases as the integrated current increases. In the present embodiment, the predetermined time Tth is set based on the test result when the integrated current is 1.0 Ah (L1).

  FIG. 8 is a test result showing an error (prediction error) between the minimum voltage Vmin calculated based on Rin and ΔImax affected by polarization as described above and the minimum voltage Vmin actually measured and tested. The plurality of test data are within the range of two solid lines in the figure. As shown in the figure, the prediction error becomes smaller as the predetermined time Tth (time threshold) is set longer. However, as the predetermined time Tth is increased, the learning opportunities for Rin and ΔImax are reduced. In view of the balance between prediction error reduction and securing of learning opportunities, the predetermined time Tth may be set based on the test result of FIG. For example, the time (5 seconds in the example of FIG. 8) corresponding to the inflection points P1 and P2 appearing on the solid line (upper limit value and lower limit value of the section estimation) in the figure may be set as the predetermined time Tth.

  FIG. 9 is a flowchart showing a procedure for determining whether or not to allow learning of Rin and ΔImax based on the time threshold (predetermined time Tth), and is repeatedly performed by the microcomputer included in the ECU 50.

  First, in step S50, it is determined whether automatic restart by the idle stop system is being performed. If it is determined that the restart has been performed, in the subsequent step S52, based on the integrated value (integrated current) of the current that has flowed to the battery 30 during the period (predetermined period) from the previous automatic stop to the current automatic stop. The predetermined time Tth (time threshold) is set. That is, as shown in FIG. 7, as the integrated current increases, the amount of change in the battery voltage immediately after the idle stop increases. Therefore, the predetermined time Tth is set longer as the integrated current increases.

  In the subsequent step S54, it is determined whether or not the time from the previous automatic stop to the start of the current restart is within a predetermined time Tth. If it is within the predetermined time Tth, learning of Rin and ΔImax calculated in steps S22 and S24 in FIG. 5 is prohibited. When learning is prohibited in this way, the minimum voltage Vmin is predicted and determined based on the previous learning value when determining whether or not to allow idle stop next time according to FIG.

  As described above, according to the present embodiment, the calculation is based on the battery voltage detected when the influence of the battery polarization is large, that is, when the time from the previous automatic stop to the start of the current restart is within the predetermined time Tth. Since learning of Rin, ΔImax (characteristic value) is prohibited, erroneous learning due to battery polarization can be avoided. Therefore, when it is determined according to FIG. 6 whether or not the idle stop is permitted next time, the prediction error of the minimum voltage Vmin can be avoided from exceeding a predetermined value, and whether or not the idle stop is permitted is determined in step S36. In the determination, it is possible to optimize the determination.

  In addition, since the time threshold value (predetermined time Tth) is variably set according to the integrated current value, the balance between the above-described prediction error reduction and learning opportunity securing can be optimized, and whether or not the learning of Rin and ΔImax is prohibited. Judgment can be performed properly.

(Second Embodiment)
In the first embodiment, learning of Rin and ΔImax is prohibited when restarting immediately after the previous automatic stop (S54: YES). On the other hand, in this embodiment, the process of step S56 in FIG. 9 is changed to step S56a shown by the one-dot chain line in the figure. That is, when restarting immediately after the previous automatic stop (S54: YES), although learning of Rin and ΔImax is permitted, a value obtained by correcting the calculated Rin and ΔImax is learned.

  Regarding the correction for the internal resistance Rin, for example, when charging was performed at the time of the previous automatic stop, the Rin calculated in step S22 of FIG. 5 should be lower than the actual value due to the influence of polarization. A correction for adding a predetermined amount to Rin may be performed. In addition, if the predetermined amount is set to a larger value as the elapsed time from the previous automatic stop to the restart is shorter, the actual internal resistance can be accurately approximated.

  In addition, if the discharge was performed at the previous automatic stop, the calculated Rin should be higher than the actual value due to the influence of polarization. Therefore, a correction for subtracting a predetermined amount from the calculated Rin should be performed. That's fine. In addition, if the predetermined amount is set to a larger value as the elapsed time from the previous automatic stop to the restart is shorter, the actual internal resistance can be accurately approximated.

  Regarding the correction for the maximum discharge current amount ΔImax, for example, when charging was performed at the time of the previous automatic stop, the calculated ΔImax should be higher than the actual value, and therefore, a predetermined amount is subtracted from the calculated ΔImax. What is necessary is just to implement the correction | amendment to perform. In addition, if the predetermined amount is set to a larger value as the elapsed time from the previous automatic stop to the restart is shorter, the actual maximum discharge current amount can be accurately approximated.

  Further, if the battery was discharged at the time of the previous automatic stop, the calculated ΔImax should be lower than the actual value, and therefore a correction for adding a predetermined amount to the calculated ΔImax may be performed. In addition, if the predetermined amount is set to a larger value as the elapsed time from the previous automatic stop to the restart is shorter, the actual maximum discharge current amount can be accurately approximated.

  In addition, regarding the correction for the maximum discharge current amount ΔImax, if ΔImax is calculated based on the equation “ΔImax = ΔV / Rin” using the corrected Rin, the correction of ΔImax is unnecessary.

  As described above, according to the present embodiment, it is possible to avoid erroneous learning of Rin and ΔImax due to the influence of polarization. Therefore, it is possible to avoid that the prediction error of the minimum voltage Vmin exceeds a predetermined value, and whether or not to allow idle stop. In determining in step S36, it is possible to optimize the determination.

(Third embodiment)
In the second embodiment, when restarting immediately after the previous automatic stop (S54: YES), the calculated Rin and ΔImax are corrected and learned. On the other hand, in the present embodiment, even when restarting immediately after the previous automatic stop (S54: YES), learning is performed without correcting the calculated Rin and ΔImax. However, the idle stop permission determination process shown in FIG. 6 is changed to the content of adding steps S34a and S34b as shown in FIG.

  That is, after the minimum voltage Vmin is predicted in step S34, in the subsequent step S34a, the time from the previous automatic stop to the start of the current restart is within the predetermined time Tth, as in step S54 of FIG. It is determined whether or not there is. If it is within the predetermined time Tth, the process proceeds to step S34b, and the predicted value of the minimum voltage Vmin calculated in step S34 is corrected.

  Regarding this correction, for example, if the battery was charged at the previous automatic stop, the minimum voltage Vmin calculated in step S34 should be higher than the actual value due to the influence of polarization. What is necessary is just to perform the correction | amendment which subtracts only fixed_quantity | quantitative_assay. In addition, if the predetermined amount is set to a larger value as the elapsed time from the previous automatic stop to the restart is shorter, the actual minimum voltage can be accurately approximated.

  Further, if the battery was discharged at the previous automatic stop, the minimum voltage Vmin calculated in step S34 should be lower than the actual value due to the influence of polarization, and therefore, a predetermined amount is added to the calculated Vmin. What is necessary is just to implement the correction | amendment to perform. In addition, if the predetermined amount is set to a larger value as the elapsed time from the previous automatic stop to the restart is shorter, the actual minimum voltage can be accurately approximated.

  As described above, according to the present embodiment, it is possible to avoid that the prediction error of the minimum voltage Vmin exceeds a predetermined value due to the influence of polarization, and in determining whether or not the idle stop is permitted in step S36, the determination is appropriate. Can be achieved.

(Fourth embodiment)
In the first to third embodiments, since the influence of polarization is large within a predetermined time Tth from the stop of charging, a measure of prohibiting learning or performing correction is performed. On the other hand, in the present embodiment, since the influence of polarization is large even within the second predetermined time from the stop of discharge, a measure for prohibiting learning or performing correction is performed.

  For example, immediately after the headlight is turned off during an idle stop or the air conditioner is turned off (immediately after the discharge is stopped), the open circuit voltage gradually increases as the polarization is gradually eliminated. (See symbol LB in FIG. 7). At this time, the rising speed of the open circuit voltage is high immediately after the end of the discharge, and the rising speed gradually decreases and converges to the value of the open circuit voltage in the state without polarization. Therefore, even when the engine is automatically restarted immediately after the battery discharge is stopped, Rin and ΔImax calculated in steps S22 and S24 are different from actual values. There is a concern that an appropriate judgment may not be made.

  Therefore, in the present embodiment, the processing in FIG. 9 described above is performed, and the processing in which the determination in step S54 in FIG. 9 is changed as follows is also performed. Alternatively, the process of FIG. 10 described above is performed, and the process in which the determination in step S34 of FIG. 10 is changed as follows is also performed. That is, it is determined whether or not automatic start has been performed within a second predetermined time after the power supply from the battery 30 to a predetermined electric load (for example, a headlight or an air conditioner having a large load) is cut off. To do.

  If it is determined that the automatic start is performed within the second predetermined time, the automatic start is regarded as being performed when the influence of the polarization is large within the second predetermined time immediately after the discharge is stopped. 9 prohibits learning in step S56, or corrects the predicted value in step S34b of FIG.

  As described above, according to the present embodiment, learning is performed by correcting the apparent characteristic value in consideration of the influence of the polarization immediately after the discharge is stopped, or the battery voltage minimum value predicted based on the apparent characteristic value is obtained. Therefore, it is possible to optimize the automatic stop permission determination.

  Note that the battery voltage decrease rate immediately after the stop of charging is greater than the battery voltage increase rate immediately after the stop of discharging. That is, the influence of polarization is greater in the case of automatic start immediately after stopping charging than in the case of automatic starting immediately after stopping discharging. In the present embodiment in view of this point, the second predetermined time immediately after the stop of the discharge, which is the subject of learning prohibition (or correction), is set to a time shorter than the predetermined time Tth immediately after the stop of charging. It is possible to promote optimization of whether or not to perform correction.

(Other embodiments)
The present invention is not limited to the description of the above embodiment, and may be modified as follows. Moreover, you may make it combine the characteristic structure of each embodiment arbitrarily, respectively.

  In each of the above embodiments, the time threshold (predetermined time Tth) is variably set according to the integrated current, but may be a fixed value set in advance.

  As shown in FIG. 7, when the engine is stopped during discharging, the influence of polarization is smaller than when the engine is stopped during charging, so the amount of change in battery voltage is small. In view of this point, it may be determined whether the engine is stopped during discharging or charging, and the time threshold (predetermined time Tth) may be variably set according to the determination result. Specifically, when the engine is stopped during discharging, the predetermined time Tth is set shorter than when the engine is stopped during charging.

  In addition, since the degree of influence of polarization changes according to the value of the internal resistance Rin and the battery temperature at that time, the time threshold (predetermined time Tth) may be variably set according to the Rin, temperature, and the like. .

  In each of the above embodiments, the maximum discharge current amount ΔImax is learned as the battery characteristic value. However, the discharge current amount (maximum value Imax) when the voltage of the battery 30 becomes the minimum value is estimated and learned. Also good. This can be realized, for example, by performing a process of calculating “Imax = Vbtm / Rin” instead of the process of step S24 of FIG.

  Here, it is considered that the discharge current of the battery 30 is small during the period before the power is supplied to the starter 40. Therefore, the process illustrated in FIG. 6 may be performed using the maximum value Imax instead of the maximum discharge current amount ΔImax.

  -The acquisition method of the open end voltage Vo before starter starting is not restricted to what was illustrated in the said 1st Embodiment. For example, the open end voltage before startup may be estimated by regression analysis based on the detected values of the voltage and current of the battery 30 when the internal combustion engine 10 is stopped.

  In the automatic starting process of the internal combustion engine 10 after the idle stop control, the starting means for applying the initial rotation to the crankshaft 12 of the internal combustion engine 10 is not limited to the starter 40. For example, a motor generator dedicated to automatic start may be provided separately from the starter 40. In this case, it is desirable to calculate the maximum discharge current amount ΔImax based on the current and voltage of the battery 30 when the internal combustion engine 10 is automatically started by the motor generator.

  The internal combustion engine is not limited to the port injection type spark ignition type internal combustion engine, and may be, for example, a cylinder injection type spark ignition type internal combustion engine. Further, it may be a compression ignition type internal combustion engine such as a diesel engine.

  DESCRIPTION OF SYMBOLS 30 ... Battery, 40 ... Starter, S26 ... Battery characteristic learning means, S36 ... Automatic stop permission determination means, Rin ... Internal resistance (characteristic value), Tth ... Predetermined time, [Delta] Imax ... Maximum discharge current amount (characteristic value).

Claims (7)

Applied to an engine system comprising: a starter for applying an initial rotation to an output shaft of an internal combustion engine; a battery for supplying power to the starter; and an idle stop system for automatically stopping and starting the internal combustion engine;
Battery characteristic learning means for calculating a characteristic value of the battery based on the battery voltage detected at the time of the automatic start, and storing the calculated characteristic value as a learning value;
When there is a request for automatic stop, automatic stop permission determination means for predicting a minimum battery voltage value at the next automatic start based on the learned value and determining whether to permit the request based on the predicted value When,
With
An automatic start control device for an internal combustion engine, wherein when the automatic start is executed within a predetermined time after the automatic stop, learning by the battery characteristic learning means at the time of the automatic start is prohibited. Applied to an engine system comprising: a starter for applying an initial rotation to an output shaft of an internal combustion engine; a battery for supplying power to the starter; and an idle stop system for automatically stopping and starting the internal combustion engine;
Battery characteristic learning means for calculating a characteristic value of the battery based on the battery voltage detected at the time of the automatic start, and storing the calculated characteristic value as a learning value;
When there is a request for automatic stop, automatic stop permission determination means for predicting a minimum battery voltage value at the next automatic start based on the learned value and determining whether to permit the request based on the predicted value When,
With
When the automatic start is performed within a predetermined time after the automatic stop, the characteristic value calculated at the automatic start is corrected and learned, or predicted based on the learned value at the automatic start An automatic start control device for an internal combustion engine, wherein the battery voltage minimum value is corrected.   When the apparent battery voltage due to the polarization of the battery changes with the recovery of the polarization from the time when the automatic stop is performed, until the amount of change per unit time of the battery voltage becomes less than a predetermined value 3. The automatic start control device for an internal combustion engine according to claim 1, wherein the time required for the internal combustion engine is the predetermined time.   4. The internal combustion engine according to claim 1, wherein an integrated value of the current flowing in the battery during a predetermined period is calculated, and the predetermined time is variably set according to the integrated value. Automatic start control device.   When the automatic start is performed within the second predetermined time after the power supply from the battery to the predetermined electric load is cut off, the learning by the battery characteristic learning means at the automatic start is prohibited. The automatic start control device for an internal combustion engine according to any one of claims 1 to 4.   When the automatic start is performed within the second predetermined time after the power supply from the battery to the predetermined electric load is cut off, the characteristic value calculated at the automatic start is corrected and learned, or 5. The automatic start control device for an internal combustion engine according to claim 1, wherein the battery voltage minimum value predicted based on the learned value at the time of automatic start is corrected.   The automatic start control device for an internal combustion engine according to claim 5 or 6, wherein the second predetermined time is set to be shorter than the predetermined time.

Images