JP2011106471A - Valve characteristic control apparatus and valve characteristic control system - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve the problem wherein a deviation of a response characteristic of a variable valve timing apparatus lowers the controllability of valve timing. <P>SOLUTION: When valve timing is held, a holding dead band appears in which the rate of change in valve timing is small even when a duty value as an operation signal to a variable valve timing apparatus is changed slightly. The holding dead band extends, and exposes individual variability with decreasing temperature. When valve timing is held, a deviation of a response characteristic of valve timing is learned according to the rate of change in valve timing under a slight change in the duty valve D from within the holding dead band to the outside of the band. <P>COPYRIGHT: (C)2011,JPO&INPIT

Description

  The present invention is a valve configured to include a fluid drive type valve characteristic variable mechanism that varies the valve characteristic of an engine valve, and a working fluid adjustment means that adjusts the state of the working fluid that acts on the valve characteristic variable mechanism. The present invention relates to a valve characteristic control device for an internal combustion engine that is applied to a characteristic variable device and controls the valve characteristic of the engine valve by operating the working fluid adjusting means, and a valve characteristic control system including the same.

  In this type of variable valve characteristic device, the valve timing of the intake valve and the exhaust valve (engine valve) of the internal combustion engine is made variable by making the relative rotation angle difference of the cam shaft relative to the output shaft of the internal combustion engine variable. There is a variable valve timing device. Here, one of the first rotating body that rotates in conjunction with the crankshaft and the second rotating body that rotates in conjunction with the camshaft is accommodated in the other, and is partitioned by these two rotating bodies. The rotational phase difference is adjusted by adjusting the state of the working fluid (working oil) in the hydraulic chamber using an oil control valve (OCV).

  By the way, the response characteristics of the variable valve timing device vary due to individual differences in OCV, changes over time, variations in flow characteristics of hydraulic fluid, and the like. In particular, when the internal combustion engine is in cold operation, the response characteristic of the variable valve timing device is low because the viscosity of the hydraulic oil is high and the frictional resistance of the variable valve timing device is high. There is a possibility that the adjustment range of the change speed of the valve timing may be narrowed.

  Therefore, conventionally, for example, as shown in Patent Document 1 below, it is also proposed to perform inching control that alternately gives a forced drive signal and a drive stop signal to the OCV during the cold operation of the internal combustion engine. ing. According to this, the responsiveness of the variable valve timing device at the time of cold start can be improved.

JP 2002-254017 A

  However, the inching control needs to adapt the repetition period of the forced drive signal and the drive stop signal, the length of each signal, etc., and there is a problem that the control design becomes complicated.

  The present invention has been made to solve the above-described problems, and an object of the present invention is to provide a valve characteristic control device and a valve characteristic that can appropriately cope with a deviation in response characteristics of the valve characteristic variable device with a simple configuration. To provide a control system.

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

The invention of claim 1 Symbol mounting includes a fluid powered valve characteristic changing mechanism for the valve characteristic of the institutional valve variable, and a working fluid regulating means for regulating the state of the working fluid to act on the valve characteristic changing mechanism The valve characteristic of the engine valve is detected in the valve characteristic control apparatus of the internal combustion engine that is applied to the valve characteristic variable device configured to control the valve characteristic of the engine valve by operating the working fluid adjusting means. Means for acquiring a detection value of the detection means, and learning means for learning a deviation amount of the response characteristic of the valve characteristic variable device based on a time change of the valve characteristic using the detection value of the detection means as an input. learning means, independently of the valve characteristics required from the operation state of the internal combustion engine, forcibly changed for operation signal the learning of the working fluid regulating means And performing the learning in a.

  The holding dead zone region, which is a region from the holding point at which the valve characteristic is held to the point at which the change speed of the valve characteristic changes rapidly with respect to the change in the operation signal, varies greatly depending on the temperature of the working fluid and the like. The inventors have found that the size of the holding dead zone has a great influence on the responsiveness of the valve characteristics of the engine valve. That is, even if the operation signal is the same, the time change of the valve characteristic varies depending on how far the operation signal is from the end of the region in the vicinity of the holding dead zone. This means that depending on the setting of the operation signal, the time variation of the valve characteristic has a strong correlation with the deviation amount of the response characteristic of the valve characteristic (deviation amount with respect to the characteristic used as a reference when controlling the valve characteristic). means. In this regard, in the above invention, since the operation signal is forcibly changed for learning, learning can be performed in a region where the correlation between the time variation of the valve characteristic and the deviation amount of the response characteristic becomes significant. For this reason, it is possible to appropriately learn the shift amount of the response characteristic.

  According to a third aspect of the present invention, in the invention according to the first or second aspect, the learning is a point in which a change speed of the valve characteristic rapidly changes with respect to a change in the operation signal from a holding point at which the valve characteristic of the engine valve is held. The amount of deviation of the response characteristic due to the deviation of the boundary of the holding dead zone area, which is the area up to, is quantified.

  As described above, the inventors have found that the size of the holding dead zone region greatly affects the responsiveness of the valve characteristics of the engine valve. Therefore, if it is possible to learn a shift in the boundary of the holding dead zone region (a shift from the boundary of the holding dead zone region in the valve characteristic serving as a reference in the control of the valve characteristic), this is a suitable representation of the shift amount of the response characteristic. .

According to a fourth aspect of the present invention, in the invention according to any one of the first to third aspects, the learning means is configured such that the operation signal is generated from a holding point where a valve characteristic of the engine valve is held. It is a value between the minimum boundary and the maximum boundary assumed from the response characteristics of the valve characteristic control device for the holding dead zone area, which is the area up to the point where the change speed of the valve characteristic with respect to the change of the valve changes suddenly. In this case, the learning is performed.

  When the operation signal is outside the holding dead zone area, the change speed of the valve characteristic increases as the distance from the holding dead zone area increases, but the changing speed hardly changes when the degree of separation becomes a certain level. At this time, the change speed of the valve characteristic is not affected by individual differences. For this reason, when the operation signal is excessively separated from the holding dead zone region, the phase difference between the response characteristic shift amount and the time change of the valve characteristic becomes small, and the response characteristic shift amount cannot be detected appropriately. In this regard, in the above-described invention, learning is performed when the operation signal changes so that the value is between the minimum boundary and the maximum boundary of the holding dead zone region assumed from the response characteristics of the valve characteristic control device. Therefore, learning can be performed in a region where the correlation between the time variation of the valve characteristic and the deviation amount of the response characteristic becomes significant. For this reason, it is possible to appropriately learn the shift amount of the response characteristic.

The invention according to claim 5 is the invention according to any one of claims 1 to 4 , wherein the learning means is lower than a temperature region in which the temperature of the working fluid converges with the operation of the internal combustion engine. The learning is performed at a temperature.

  The variation in the boundary of the holding dead zone region between individuals tends to become more prominent as the temperature of the working fluid is lower. In this regard, in the above invention, the learning is performed at a temperature lower than the temperature range where the temperature of the working fluid rises and converges with the operation of the internal combustion engine, so that the variation in the boundary of the holding dead zone region becomes apparent. Can learn.

The invention according to claim 5 may be characterized in that the learning is performed when the difference between the temperature of the working fluid and the temperature of the outside air is not more than a predetermined value. Thereby, learning can be performed at a temperature at which the variation of the boundary of the holding dead zone region becomes more apparent.

The invention according to claim 6 is the invention according to any one of claims 1 to 5 , wherein the learning means changes the actual valve characteristic to the target value in accordance with a change in the target value of the valve characteristic. The learning is performed when feedback control is performed.

  Even if the same operation signal is given, the rate of change of the valve characteristic can be different depending on the individual difference of the valve characteristic. Here, when the operation signal is directly determined and given to the working fluid adjusting means during learning, the change speed may be excessively increased for an individual having high response characteristics. On the other hand, if the operation signal is set so as to avoid such a situation, the difference in valve characteristics between those having low response characteristics may not be sufficiently identified. In this regard, in the above-described invention, learning can be performed while avoiding excessive changes in the valve characteristics of the engine valve by performing learning when feedback control of the actual valve characteristics to the target value.

The invention according to claim 7 is the invention according to any one of claims 1 to 6 , wherein the valve characteristic of the engine valve is controlled so as to compensate for the deviation amount learned by the learning means. The apparatus further comprises operation signal setting means for setting an operation signal of the working fluid adjusting means.

  In the above invention, since the operation signal for controlling the valve characteristic is set so as to compensate for the learned deviation amount, the controllability of the valve characteristic can be improved.

The invention according to claim 8 is the invention according to claim 7 , wherein the operation signal setting means takes into account the temperature of the working fluid when setting the operation signal to compensate for the learned shift amount. It is characterized by.

  The deviation amount of the response characteristic can vary depending on the temperature of the working fluid. For this reason, when there is a difference between the temperature when the deviation amount is learned and the temperature when the operation signal is set by the operation signal setting means, the learned deviation amount is equal to the current deviation amount. May be different. In this regard, in the above invention, by taking into account the temperature of the working fluid, the current deviation amount is grasped based on the fluctuation amount of the deviation amount of the response characteristic due to the difference between the learning temperature and the current temperature. The operation signal can be set so as to compensate.

The invention according to claim 9 is the invention according to claim 7 or 8 , wherein the operation signal setting means outputs the output of the working fluid adjustment means when setting the operation signal to compensate for the learned deviation amount. The pressure of the working fluid is taken into consideration.

  The deviation amount of the response characteristic can vary depending on the pressure of the working fluid. For this reason, when there is a difference between the pressure when the deviation amount is learned and the pressure when the operation signal is set by the operation signal setting means, the learned deviation amount is the current deviation amount. May be different. In this regard, in the above invention, by taking into account the pressure of the working fluid, the current deviation amount is grasped based on the fluctuation amount of the deviation amount of the response characteristic due to the difference between the learning pressure and the current pressure. The operation signal can be set so as to compensate.

Invention of claim 1 0, wherein, in the invention of claim 9, wherein said working fluid control means is configured to include a pump for engine-driven type, the operation signal setting means, the pressure of the working fluid, engine It is characterized by grasping based on the detected value of the rotational speed.

  When the working fluid adjustment means is configured to include an engine-driven pump, the pressure of the working fluid output from the adjustment means depends on the engine rotation speed. In the above invention, paying attention to this point, the pressure of the working fluid can be grasped based on the detected value of the engine rotational speed.

The invention of claim 1 1, wherein, in the invention according to any one of claims 1 to 1 0, wherein the learning means, the working fluid is achieved thermal equilibrium with the surrounding If it is determined that the learning is performed, the learning is performed.

  Since there are various restrictions in the vicinity of the working fluid adjusting means and the valve characteristic variable mechanism, it is assumed that a means for detecting the temperature of the working fluid cannot be provided. In such a case, it is desired to estimate the temperature of the working fluid adjusting means and the working fluid temperature of the variable valve characteristic mechanism. However, there is a concern that this estimation error may become so large that the deviation amount of the response characteristic may fluctuate. The In this case, even if the deviation amount of the response characteristic is learned, it is difficult to accurately and effectively utilize the learning result because it cannot be accurately grasped in which temperature region of the working fluid. There is a risk. In this respect, in the above-described invention, learning is performed when it is determined that the working fluid has achieved a thermal equilibrium state with the surroundings, so that the working fluid temperature and parameters of the working fluid are determined according to the parameters having the phases. The accuracy of temperature estimation can be improved.

Invention of claim 1 wherein, in the invention described in any one of claims 1 to 1 1, wherein the valve characteristic is intended to be quantified by one-dimensional parameter, the working fluid adjustment means The one-dimensional parameter is displaced in one direction or the other direction depending on whether the operation signal is on either side of the holding point, and the learning unit is configured to detect the holding point of the working fluid adjusting unit . The learning is performed separately for each of both sides.

  How much the operation signal should be changed in order to displace the parameter stagnant at a predetermined value in one direction depends on the degree of separation in one direction with respect to the current operation signal value holding point. Also, how much the operation signal should be changed in order to displace the parameter stagnant at a predetermined value in the other direction depends on the degree of separation in the other direction with respect to the current operation signal value holding point. To do. For this reason, there are two amounts of deviation of the response characteristic of the valve characteristic, that is, when the valve characteristic is displaced in one direction and when the valve characteristic is displaced in the other direction. In this regard, in the above-described invention, since these two shift amounts are learned separately, it is possible to appropriately cope with the shift in response characteristics.

The invention of claim 1 3, wherein, in the invention according to any one of claims 1 to 1 2, wherein the valve characteristic changing mechanism, the relative rotation angle difference between the cam shaft to the output shaft of the internal combustion engine The valve timing of the engine valve is made variable by making it variable.

The invention of claim 1 4, wherein, in the invention according to any one of claims 1 to 1 3, wherein the learning means, the change speed and the valve characteristic of the valve characteristics based on the detection value of said detection means The learning is performed based on at least one of the time integrated values of the quantified values.

  In the above invention, it is possible to perform the learning by appropriately quantifying the time change of the valve characteristic. In particular, if an integral value is used, the above time change can be quantified while suitably removing the influence of slight fluctuations in the detection value of the detection means.

The invention of claim 1 5, wherein, in the invention according to any one of claims 1 to 1 4, wherein the learning means, based on the difference between the target value of the detection value and the valve characteristic of the detecting means and the The learning is performed while quantifying the time change of the valve characteristic.

  In the above invention, it is possible to perform the learning by appropriately quantifying the time change of the valve characteristic. In particular, in the above-described invention, it is easy to remove the influence of the variation in the detection value according to the degree of change in the target value by adding the target value.

The invention of claim 1 6, wherein, in the invention according to any one of claims 1 to 1 5, wherein the learning means, the detecting means to the target value in accordance with a change in the target value of the valve characteristic The learning is performed when the detected value of the control signal is feedback-controlled, and the learning is performed while quantifying the time change of the valve characteristic based on the time integral value of the operation signal used for the feedback control. And

  Even if the same operation signal is given, the rate of change of the valve characteristic can be different depending on the individual difference of the valve characteristic. Here, when the operation signal is directly determined and given to the working fluid adjusting means during learning, the change speed may be excessively increased for an individual having high response characteristics. On the other hand, if the operation signal is set so as to avoid such a situation, the difference in valve characteristics between those having low response characteristics may not be sufficiently identified. In this regard, in the above-described invention, learning can be performed while avoiding excessive changes in the valve characteristics of the engine valve by performing learning when feedback control of the actual valve characteristics to the target value.

  By the way, when feedback control is performed, since the operation signal is not constant, it is difficult to associate which value of the change value of the detected value of the valve characteristic corresponds to the value of the operation signal. There is a concern that it is difficult to associate the quantity. In this regard, in the above invention, by quantifying the time change of the valve characteristic based on the time integral value of the operation signal, it is possible to appropriately associate this with the deviation amount of the response characteristic.

The invention according to claim 17 is the invention according to claim 16 , wherein the operation signal is a duty signal, and the learning means supplies power for generating the operation signal when learning the deviation amount. In consideration of the voltage value of

  When the voltage of the power supply means changes, the amount of energy input to the working fluid adjustment means changes even with the same duty. For this reason, the operation signal required when performing feedback control to the target value can vary depending on the voltage. Therefore, the time integral value can also change depending on the voltage of the power supply means. In the above invention, in view of this point, learning can be performed while appropriately removing the shift due to the change in the voltage value of the power supply means by learning the shift amount in consideration of the voltage of the power supply means.

The invention according to claim 18 is a valve characteristic control system comprising the valve characteristic control device according to any one of claims 1 to 17 and the valve characteristic variable device.

The figure which shows the whole structure of the control system concerning 1st Embodiment. The figure which shows the relationship between the operation signal of OCV concerning the embodiment, and cam angular displacement speed. 6 is a flowchart showing a processing procedure related to control of valve timing according to the embodiment. 6 is a flowchart showing a processing procedure of retained duty learning control according to the embodiment; The figure which shows the relationship between the response characteristic of a variable valve timing apparatus, and temperature. The figure which shows the holding dead zone area | region dependence of the response characteristic of a variable valve timing apparatus. 4 is a time chart showing a duty value setting method for learning according to the embodiment. 6 is a flowchart showing a processing procedure of response characteristic learning control according to the embodiment; The block diagram which shows the calculation process of the correction amount based on the learning result concerning the embodiment. The time chart which shows the improvement effect of the controllability by the said correction amount. 9 is a flowchart showing a processing procedure of response characteristic learning control according to the second embodiment; The time chart which shows the test pattern of the embodiment. 10 is a flowchart showing a processing procedure of response characteristic learning control according to the third embodiment; 10 is a flowchart showing a processing procedure of response characteristic learning control according to the fourth embodiment; 10 is a flowchart showing a processing procedure of response characteristic learning control according to the fifth embodiment; 10 is a flowchart showing a processing procedure of response characteristic learning control according to the sixth embodiment;

(First embodiment)
A first embodiment in which a valve characteristic control device and a valve characteristic control system according to the present invention are applied to a variable valve timing device and a control system of a gasoline engine will be described below with reference to the drawings.

  FIG. 1 shows an overall configuration of a control system according to the present embodiment.

  As shown in the figure, the power of the crankshaft 10 is transmitted to the camshaft 14 via the belt 12 and the variable valve timing mechanism 20. The variable valve timing mechanism 20 includes a first rotating body 21 that is mechanically connected to the crankshaft 10 and a second rotating body 22 that is mechanically connected to the camshaft 14. In the present embodiment, the second rotating body 22 includes a plurality of protrusions 22 a, and the second rotating body 22 is accommodated in the first rotating body 21. Then, the retard angle for retarding the relative rotation angle (rotation angle difference) of the cam shaft 14 with respect to the crankshaft 10 by the projection 22a of the second rotation body 22 and the inner wall of the first rotation body 21. A chamber 23 and an advance chamber 24 for advancing the rotation angle difference are defined. The variable valve timing mechanism 20 further fixes the first rotating body 21 and the second rotating body 22 at a rotation phase difference (maximum retard angle position) at which the volume of the retard chamber 23 is maximized. A lock mechanism 25 is provided.

  The variable valve timing mechanism 20 is hydraulically driven by the inflow and outflow of incompressible working fluid (working oil) between the retard chamber 23 and the advance chamber 24. This hydraulic oil inflow / outflow is a hydraulically driven actuator that is adjusted by an oil control valve (OCV30).

  The OCV 30 supplies the hydraulic oil discharged from the engine-driven pump P to which driving force is applied from the crankshaft 10 via the supply path 31 and the retard path 32 or the advance path 33 to the retard chamber 23 or the advance chamber. 24. Further, the OCV 30 causes hydraulic oil to flow from the retard chamber 23 or the advance chamber 24 to the oil pan OP via the retard path 32 or the advance path 33 and the discharge path 34. The flow path area between the retard path 32 or the advance path 33 and the supply path 31 or the discharge path 34 is adjusted by the spool 35. That is, the spool 35 is pushed to the left side in the drawing by the spring 36 and a force toward the right side in the drawing is applied by the electromagnetic solenoid 37. For this reason, it is possible to manipulate the displacement amount of the spool 35 by giving an operation signal to the electromagnetic solenoid 37 and adjusting the duty of this operation signal.

  Control of the rotation angle difference by the operation of the OCV 30 is performed by an electronic control unit (ECU 40). The ECU 40 is mainly composed of a microcomputer. The crank angle sensor 50 detects the rotation angle of the crankshaft 10, the cam angle sensor 52 detects the rotation angle of the camshaft 14, the water temperature sensor 54 detects the cooling water temperature of the gasoline engine, and the air flow detects the intake air amount. The detected values of various operating states of the internal combustion engine such as the meter 56 are captured. Then, various calculations are performed based on the detected values, and various actuators of the internal combustion engine such as the OCV 30 are operated based on the calculation results.

  The ECU 40 is provided with various memories such as a constant memory holding device 42 in order to store and hold data used for the various calculations. Here, the constant storage holding device 42 is a storage device that always holds data regardless of the main connection state (the state of the power switch SW) between the ECU 40 and the battery B as the power supply means. As this constant memory holding device 42, for example, a backup memory that is always in a power supply state regardless of the state of main connection between the ECU 40 and the battery B, or a non-volatile memory (EEPROM, etc.) that holds data regardless of whether power is supplied or not. and so on.

  Hereinafter, the control of the rotational phase difference by the ECU 40 will be described in detail.

  When the force by which the spring 36 pushes the spool in the left direction in the figure is larger than the force by which the magnetic field of the electromagnetic solenoid 37 displaces the spool 35 in the reverse direction, the spool 35 is displaced in the left direction in the figure. When the spool 35 is displaced to the left from the illustrated position, oil is supplied from the hydraulic pump to the retard chamber 23 via the supply path 31 and the retard path 32, and from the advance chamber 24 to the advance path 33. The oil is discharged to the oil pan OP through the discharge path 34. Thereby, the 2nd rotary body 22 rotates counterclockwise in the figure.

  On the other hand, when the force by which the magnetic field of the electromagnetic solenoid 37 displaces the spool 35 in the right direction is larger than the force by which the spring 36 pushes the spool in the left direction in the figure, the spool 35 is displaced in the right direction in the figure. When the spool 35 is displaced to the right from the illustrated position, oil is supplied from the hydraulic pump to the advance chamber 24 via the supply path 31 and the advance path 33, and from the retard chamber 23 to the retard path 32. The oil is discharged to the oil pan OP through the discharge path 34. Thereby, the 2nd rotary body 22 rotates clockwise in the figure.

  However, as shown in the figure, when the spool 35 is in a position closing the retard path 32 and the advance path 33, the flow of oil between the retard chamber 23 and the advance chamber 24 is stopped, and the rotation angle The difference is retained.

  In the ECU 40, the electromagnetic solenoid 37 of the OCV 30 is energized to operate the opening of the spool 35 and control the rotation angle difference. In particular, in the present embodiment, the duty ratio control (duty control) for adjusting the ratio of the on time (or off time) to one cycle when the operation signal is changed in a binary and periodic manner is energized to the electromagnetic solenoid 37. ). FIG. 2 shows the relationship between the duty value of the operation signal for the electromagnetic solenoid 37 and the displacement speed of the camshaft 14 relative to the crankshaft 10.

  As shown in the drawing, when the duty value is the value D0, the displacement speed becomes zero. In other words, the rotational phase difference is maintained when the duty value is the value D0. On the other hand, when the duty value is smaller than the value D0, the camshaft 14 is displaced toward the retard side, and the retard side displacement speed increases as the duty value decreases. On the other hand, when the duty value is larger than the value D0, the cam shaft 14 is displaced to the advance side, and the advance side displacement speed is increased as the duty value is increased.

  Therefore, by learning “D0”, which is a duty value for holding the rotational phase difference, as the retained duty value and performing feedback control of the rotational angle difference to the target value based on the retained duty value, the rotational angle difference is obtained. Can be appropriately controlled to the target value.

  FIG. 3 shows a processing procedure of the feedback control of the rotation angle difference according to the present embodiment. This process is repeatedly executed by the ECU 40, for example, at a predetermined cycle.

  In this series of processes, first, in step S10, based on parameters that determine the operating state of the internal combustion engine, such as the rotational speed of the crankshaft 10 and the intake air amount, the target value of the rotational angle difference of the camshaft 14 with respect to the crankshaft 10 is obtained. A target advance value VCTa is calculated.

  In the subsequent step S12, an actual advance angle value VCTr, which is an actual rotation angle difference of the camshaft 14 with respect to the crankshaft 10, is calculated based on the detected value of the crank angle sensor 50 and the detected value of the cam angle sensor 52. In the subsequent step S14, it is determined whether or not the absolute value of the difference between the actual advance value VCTr and the target advance value VCTa is equal to or greater than a predetermined value α. The predetermined value α defines a threshold value for performing feedback control during transition based on the difference between the actual advance value VCTr and the target advance value VCTa.

  When a positive determination is made in step S14, the actual advance value VCTr is feedback-controlled to the target advance value VCTa. Here, first, in step S16, a proportional term FBP and a differential term FBD based on the deviation Δ of the target advance value VCTa with respect to the actual advance value VCTr are calculated. In the subsequent step S18, the duty value D of the operation signal is calculated. Here, the duty value D is defined by the ratio of the on time to the on / off period. The duty value D is calculated by adding a sum of a proportional term FBP and a differential term FBD and a correction amount OFD described later and a correction coefficient K and a holding duty value KD described later. Here, the correction coefficient K is for compensating for the change in the voltage VB of the battery B. That is, a change in the amount of energy input to the OCV 30 resulting from a change in the voltage of the battery B from a reference value (for example, “14V”) is compensated, and the same energy is input regardless of the voltage VB of the battery B. When the duty value D is calculated in this way, the OCV 30 is operated based on this in step S20.

  When a negative determination is made in step S14 or when the process of step S20 is completed, this series of processes is temporarily terminated.

  FIG. 4 shows the learning control procedure for the retained duty value KD. This process is repeatedly executed by the ECU 40, for example, at a predetermined cycle.

  In this series of processing, first, in step S30, it is determined whether or not the target advance value VCTa and the actual advance value VCTr are in a stable state for a predetermined time or longer. This process determines whether the feedback control has converged. Here, it is possible to determine that each parameter is stable depending on whether or not these fluctuation amounts are equal to or less than a predetermined value. If a positive determination is made in step S30, the process proceeds to step S32.

  In step S32, it is determined whether or not the absolute value of the deviation Δ of the target advance value VCTa with respect to the actual advance value VCTr is equal to or greater than a predetermined value β. This process determines whether or not a steady difference has occurred between the actual advance value VCTr and the target advance value VCTa by the feedback control. Here, the predetermined value β is set to a value for determining whether or not the above-described steady difference has occurred. If an affirmative determination is made in step S32, it is determined that a steady difference has occurred between the actual advance value VCTr and the target advance value VCTa by feedback control, and the process proceeds to step S34.

  In step S34, the retained duty value D is updated. That is, when a steady difference occurs even though the feedback control shown in FIG. 3 is performed, it is considered that the hold duty value KD is deviated from an appropriate value. Update the Duty value. Here, the retained duty value is updated to the current duty value D. As a result, the difference between the target advance value VCTa and the actual advance value VCTr is reduced. When the duty value D determined by updating the held duty value KD is excessive, the duty value D is updated by the feedback control shown in FIG.

  On the other hand, if a negative determination is made in step S32, the duty value D is set as the retained duty value D in step S36. When a negative determination is made at step S30 or when the processes at steps S34 and S36 are completed, this series of processes is temporarily terminated.

  By the way, the relationship (response characteristic) between the duty value D and the actual advance value VCTr shown in FIG. 2 fluctuates due to individual differences, changes over time, temperature, and the like. In particular, variation in response characteristics is noticeably caused by the influence of temperature. Hereinafter, this will be described with reference to FIG.

  On the left side of FIG. 5, an example of response characteristics of a variable valve timing device configured to include the variable valve timing mechanism 20 and the OCV 30 is shown. Here, both a1 and a2 have a very small change rate of the actual advance value VCTr even if the duty value D is slightly changed in a state where the actual advance value VCTr is once held at the hold duty value KD. A certain holding dead zone region is shown. As shown in the figure, when the duty value D is changed from the held duty value KD, there is a point where the change rate of the actual advance value VCTr changes suddenly, but in the region from the held duty value KD to the point where the sudden change occurs. The change rate of the actual advance value VCTr becomes very small. Specifically, a2 is a holding dead zone region on the retard side, and a1 is a holding dead zone region on the advance side. On the other hand, b1 and b2 indicate regions in which the change speed of the actual advance value VCTr changes remarkably according to the change of the duty value D. Specifically, b2 is a retarded side region, and b1 is an advanced side region. Further, c1 and c2 indicate upper limit speeds in a region (dead zone) where the change speed of the actual advance value VCTr hardly changes even when the duty value D is changed. Specifically, c2 is the retard side speed, and c1 is the advance side speed.

  The right side of FIG. 5 shows the relationship between the width of the holding dead zone region and the hydraulic oil temperature. In the figure, the one-dot chain line indicates the width of the holding dead zone for the highest response characteristic among the variable valve timing devices shipped as products, and the solid line indicates the holding dead band for the lowest response characteristic. Indicates the width. As shown in the drawing, the width of the holding dead zone region increases as the temperature of the hydraulic oil decreases, and the variation width of the width of the holding dead zone region with respect to the temperature change is very large. Furthermore, in the temperature range (70 to hundreds of degrees) where the temperature of the hydraulic oil rises and converges due to the operation of the gasoline engine, the individual difference in the width of the holding dead zone region is very small, but is higher than the above temperature range. As the temperature becomes lower, individual differences in the width of the holding dead zone region become more prominent.

  Thus, it can be seen that the variation of the width of the holding dead zone region with respect to the change of the temperature of the hydraulic oil is large, and the variation due to the individual difference can be very large. Here, in relation to the proportional term FBP and the differential term FBD of the feedback control in FIG. 3, the relationship with the deviation Δ is determined in consideration of the holding dead zone region. However, if this holding dead zone region changes with temperature and the variation of individual differences is very large, the amount of deviation between the response characteristic serving as a reference assumed when controlling the actual advance value VCTr and the actual response characteristic May increase, and as a result, controllability may be reduced.

  FIG. 6 shows that the overshoot amount or the undershoot amount is within an allowable range when the target advance value VCTa is changed stepwise for a variable valve timing device having the highest response characteristic at a predetermined temperature. In the case where the control is adapted to suppress hunting, the influence due to the expansion of the holding dead zone region is shown. Here, the circle indicates the arrival time of “90%” when the target advance value VCTa is changed in a step shape from “10 ° CA” to “35 ° CA”, and the triangle indicates the target advance value The arrival time of “90%” when VCTa is changed stepwise from “10 ° CA” to “17 ° CA” is shown. As illustrated, when the width of the holding dead zone region changes, the arrival time changes significantly. For this reason, as shown in FIG. 5 above, when the holding dead zone region has a property of largely changing according to the temperature, the controllability particularly when the temperature of the hydraulic oil is low is lowered.

  Therefore, in this embodiment, the shift amount of the response characteristic caused by the shift of the boundary of the width of the holding dead zone region is quantified and learned, and the correction amount OFD shown in FIG. 3 is calculated based on this. Hereinafter, this will be described with reference to FIG.

  FIG. 7C shows a variable valve timing device shipped with a product having the highest response characteristic by a one-dot chain line, and a thing having the lowest response characteristic by a solid line. As shown in the figure, the deviation of the change rate of the actual advance value VCTr becomes noticeable when the duty value D is slightly changed outside the holding dead zone due to the deviation of the boundary of the holding dead zone. For this reason, for example, as shown in FIG. 7C, the duty value is changed by changing the Duty value to a value between the shortest boundary and the longest boundary of the holding dead zone in the variable valve timing device. It is considered that the shift amount of the response characteristic due to the shift of the boundary of the dead zone can be detected.

  Therefore, as shown in FIG. 7A, a test pattern for changing the duty value in the above-described manner is set in advance. Then, as shown in FIG. 7 (b), the time change of the actual advance value VCTr at this time is detected, and based on this, the actual response characteristic with respect to the response characteristic serving as a reference in controlling the actual advance value VCTr is determined. Quantify and learn the amount of deviation.

  FIG. 8 shows the procedure of the learning process of the deviation amount of the response characteristic according to the present embodiment. This process is repeatedly executed by the ECU 40, for example, at a predetermined cycle.

In this series of processing, first, in step S40, it is determined whether or not a learning execution condition is satisfied. Here, examples of the learning execution condition include the following.
Condition a. The cooling water temperature detected by the water temperature sensor 54 is about a specified temperature THW0 (≦ 0 ° C.).
Condition b. The estimated value of the hydraulic oil temperature is equivalent to the coolant temperature.
Condition c. The engine stop time immediately before the operation of the engine is a predetermined time Tr or more. Here, the predetermined time Tr is set to be longer than the time when the hydraulic oil is assumed to be in thermal equilibrium with the surroundings when the engine is stopped.
Condition d. The rotational speed is about the predetermined speed NE0.

  The above conditions a to c are for determining whether or not the hydraulic oil has achieved a thermal equilibrium state with the surroundings. This is for determining whether or not the estimation accuracy of the temperature of the hydraulic oil is high. That is, in the estimation of the temperature of the hydraulic oil based on a well-known method, an error of about “± several degrees to twenty degrees” may normally occur. However, as shown in FIG. The response characteristics can vary greatly. For this reason, in order to accurately estimate the temperature of the hydraulic oil in the variable valve timing mechanism 20 and the OCV 30, it is necessary to realize the above thermal equilibrium state. If these conditions are satisfied, the temperature of the hydraulic oil can be expressed with high accuracy by the temperature of the cooling water.

  In the subsequent step S42, the test pattern of the duty value D shown in FIG. 7A is input. In step S44, the actual advance value VCTr shown in FIG. 7B is detected. In step S46, a change speed ΔVb when the actual advance value VCTr shifts to the advance side and a change speed ΔVa when it shifts to the retard side are calculated. Here, for example, an average speed may be calculated in a period in which the duty value is a value on the advance side or the retard side. In the subsequent step S48, is at least one of the conditions that the absolute value of the change speed ΔVb is smaller than the absolute value of the reference speed ΔVb0 and that the absolute value of the change speed ΔVa is smaller than the absolute value of the change speed ΔVa0 satisfied? Judge whether or not.

  Here, the reference speeds ΔVb0 and ΔVa0 are speeds at the highest of the response characteristics of the variable valve timing device shipped as a product. This is related to the fact that, in the present embodiment, the highest response characteristic of the variable valve timing device shipped as a product is adopted as a reference response characteristic when controlling the actual advance value VCTr. Yes. The reason why the characteristic with the highest responsiveness is used as the standard characteristic is that priority is given to adaptation that suppresses control hunting. The reference speeds ΔVb0 and ΔVa0 are stored in the constant storage holding device 42.

  If an affirmative determination is made in step S48, the actual response characteristic is considered to be lower than the reference response characteristic used when controlling the actual advance value VCTr, and the process proceeds to step S50. In step S50, a learning value ODFb for compensating for the shift amount of the response characteristic is calculated. Here, the learning value OFDb / OFDa is a learning value on the advance side, and the learning value OFDa is a learning value on the retard side. In subsequent step S52, the learning value OFDb / OFDa is subjected to guard processing so that the learning value OFDb / OFDa does not become an excessively large value. Then, the learned value ODFb / OFDa subjected to the guard process is stored in the constant memory holding device 42.

  When a negative determination is made in steps S40 and S48, or when the process of step S52 is completed, this series of processes is temporarily terminated.

  By calculating the correction amount OFD by the process shown in FIG. 9 using the learned values OFDb / OFDa learned in this way, the retained duty value KD can be corrected so as to compensate for the deviation amount of the response characteristic. Here, the correction amount OFD is calculated by correcting the learning value OFDb / OFDa with the correction coefficients K1 and K2. Here, the correction coefficient K1 is for compensating for a variation in response characteristics due to a change in the rotational speed. That is, since the pump P is engine driven, the discharge pressure of the pump P depends on the rotational speed. For this reason, the correction coefficient K1 is determined so that the absolute value of the correction amount OFD increases as the rotational speed decreases. On the other hand, the correction coefficient K2 is for compensating for fluctuations in response characteristics due to changes in hydraulic oil temperature. Here, the correction coefficient K2 is determined so that the absolute value of the correction amount OFD increases as the temperature of the hydraulic oil decreases.

  The correction amount OFD calculated in this way corresponds to the learning value OFDb / OFDa corresponding to the displacement direction of the actual advance value VCTr obtained by the process shown in FIG. Then, by using this correction amount OFD for setting the duty value D, it is possible to obtain the same effect as when the holding dead zone region is reduced.

  FIG. 10 shows the effect of using the correction amount OFD.

  In FIG. 10A, a thick line indicates a reference characteristic, and a thin line indicates an actual characteristic. In this case, the actual speed is insufficient with respect to the speed expected by setting the duty value D indicated by the two-dot chain line. Here, according to the difference between the duty value determined from the actual speed and the reference characteristics and the current actual duty value D (two-dot chain line), the duty value for compensating for the shortage of the speed is simply set. it can.

  FIG. 10B shows the followability of the actual advance value VCTr to the target advance value VCTa when the learned value determined in this manner is not used, and FIG. 10C shows the case where the learned value is used. This shows the followability of the actual advance value VCTr to the target advance value VCTa. As shown in the figure, by using the learning value, the followability of the actual advance value VCTr to the target advance value VCTa is improved.

  According to the embodiment described in detail above, the following effects can be obtained.

  (1) Learning the deviation amount of the response characteristic of the variable valve timing device based on the time change of the actual advance value VCTr when the duty value D is changed from the inside of the holding dead zone of the actual advance value VCTr to the outside of the area. did. As a result, it is possible to appropriately cope with a shift in response characteristics.

  (2) Learning was performed by forcibly changing the duty value D to the learning (test pattern) independently of the target advance value VCTa required from the operating state of the gasoline engine. As a result, learning can be performed in a region where the correlation between the temporal change of the actual advance value VCTr and the deviation amount of the response characteristic becomes significant.

  (3) The shift amount of the response characteristic due to the shift of the boundary of the holding dead zone region was quantified as the learning value OFDb / OFDa. As a result, the learning value OFDb / OFDa can be appropriately expressed as a deviation amount of the response characteristic.

  (4) Learning was performed when the temperature of the hydraulic oil was about the specified temperature THW0 (≦ 0 ° C.). Thereby, learning can be performed at a temperature at which the variation of the boundary of the holding dead zone region becomes apparent.

  (5) The duty value D is set based on the correction amount OFD so as to control the actual advance value VCTr so as to compensate the learned shift amount. Thereby, the controllability of the valve characteristics can be improved.

  (6) When setting the correction amount OFD, the temperature of the hydraulic oil was taken into account. As a result, the current deviation amount is grasped and compensated for in consideration that the deviation amount of the response characteristic varies due to the difference between the learning temperature (specified temperature THW0) and the current hydraulic oil temperature. Thus, the duty value D can be set.

  (7) When setting the correction amount OFD, the pressure of the hydraulic oil discharged from the pump P was taken into account. As a result, the current deviation amount is grasped in consideration of the fluctuation amount of the response characteristic due to the difference between the learning pressure and the current pressure, and the duty value D is set to compensate for this. can do.

  (8) Learning was performed when it was determined that the hydraulic oil had achieved a thermal equilibrium with the surroundings. Thereby, learning can be performed only when the estimation accuracy of the temperature of the hydraulic oil can be maintained high in a situation where the temperature of the hydraulic oil inside the variable valve timing mechanism 20 and the OCV 30 cannot be directly detected. .

  (9) The learning value OFDb / OFDa was calculated for each of the advance side and the retard side. Thereby, it is possible to appropriately cope with a shift in response characteristics.

  (10) Based on the change rate of the actual advance value VCTr, the learning value OFDb / OFDa was calculated while quantifying the temporal change of the actual advance value VCTr. Thereby, learning can be performed by appropriately quantifying the time change.

(Second Embodiment)
Hereinafter, the second embodiment will be described with reference to the drawings with a focus on differences from the first embodiment.

  FIG. 11 shows the procedure of the learning process of the deviation amount of the response characteristic according to the present embodiment. This process is repeated by the ECU 40, for example, at a predetermined cycle. In FIG. 11, the same steps as those shown in FIG. 8 are given the same step numbers for the sake of convenience.

  In this series of processes, in step S42a, the target advance value VCTa is defined by the learning test pattern. FIG. 12 (a) shows this test pattern, FIG. 12 (b) shows the transition of the duty value D when feedback control of the actual advance value VCTr to the target advance value VCTa, and FIG. 12 (c). The transition of the actual advance value VCTr is shown. As shown in the figure, the target advance value VCTa is once forcedly changed to the advance side or the retard side, and then returned to the original state after a predetermined period, so that the Duty value D is maintained from the holding dead zone area to the holding dead zone area. It can be changed to both the advance side and the retard side.

  Moreover, when the target advance value VCTa is defined in this way, even if the response characteristics of the actual variable valve timing device are high, it is possible to avoid the actual advance value VCTr from changing excessively. it can. The target advance value VCTa is preferably set so that the maximum value of the change amount of the Duty value D by the feedback control is about the change amount in the first embodiment.

  According to this embodiment described above, in addition to the effects (1) to (10) of the first embodiment, the following effects can be obtained.

  (11) The target advance value VCTa is defined by a test pattern, and the actual advance value VCTr is feedback-controlled to the target advance value VCTa. Thus, learning can be performed while avoiding an excessive change in the actual advance value VCTr.

(Third embodiment)
Hereinafter, the third embodiment will be described with reference to the drawings with a focus on differences from the second embodiment.

  FIG. 13 shows the procedure of the learning process of the response characteristic deviation amount according to the present embodiment. This process is repeated by the ECU 40, for example, at a predetermined cycle. In FIG. 13, the same steps as those shown in FIG. 11 are given the same step numbers for the sake of convenience.

  In this series of processing, in step S46b, the time integration value IVb when the actual advance value VCTr is shifted to the advance side and the time integration value IVa when it is shifted to the retard side are calculated. Here, the integral value is an integral value with the actual advance value VCTr before the test pattern is input as the reference value 0. In step S48b, at least one of the absolute value of the time integral value IVb being smaller than the absolute value of the reference integral value IVb0 and the absolute value of the time integral value IVa being smaller than the absolute value of the reference integral value IVa0. It is determined whether the condition is satisfied. Here, the reference integral values IVb0 and IVa0 are integral values of the actual advance value VCTr in the highest response characteristic of the variable valve timing device. Since the absolute value of the change speed of the actual advance value VCTr increases as the response characteristic increases, it is assumed that the integral value of the actual advance value VCTr also increases. For this reason, when it is smaller than the reference integral values IVb0 and IVa0, it is considered to be lower than the reference characteristic. For this reason, when a positive determination is made in step S48b, the process proceeds to step S50.

  In this way, by quantifying the time change of the actual advance angle value VCTr when the duty value D is changed from the inside of the holding dead zone region to the outside of the region by the integrated value, the above FIG. The time change can be quantified while suitably eliminating the influence of the minute time change of the actual advance value VCTr as shown in FIG.

(Fourth embodiment)
Hereinafter, the fourth embodiment will be described with reference to the drawings with a focus on differences from the second embodiment.

  FIG. 14 shows the procedure of the learning process of the deviation amount of the response characteristic according to the present embodiment. This process is repeated by the ECU 40, for example, at a predetermined cycle. In FIG. 14, the same steps as those shown in FIG. 11 are given the same step numbers for the sake of convenience.

  In this series of processing, in step S46c, when the actual advance value VCTr is shifted to the advance side, the change speed ΔVb of the difference between the actual advance value VCTr and the target advance value VCTr and the actual advance value VCTr are changed. A change speed ΔVa of a difference between the actual advance value VCTr and the target advance value VCTa at the time of shifting to the retard side is calculated. That is, in the present embodiment, the time change of the actual advance value VCTr is quantified based on the change rate of the difference between the target advance value VCTr and the actual advance value VCTr.

(Fifth embodiment)
Hereinafter, the fifth embodiment will be described with reference to the drawings with a focus on differences from the second embodiment.

  FIG. 15 shows the procedure of the learning process of the deviation amount of the response characteristic according to the present embodiment. This process is repeated by the ECU 40, for example, at a predetermined cycle. In FIG. 15, the same steps as those shown in FIG. 11 are given the same step numbers for the sake of convenience.

  In this series of processing, in step S46d, the time integral value IDb of the duty value when the actual advance value VCTr is shifted to the advance side and the time integral value IDa of the duty value when the shift is made to the retard side are calculated. To do. Here, the integral value is an integral value in which the hold duty value KD before the test pattern is input is set to the reference value 0. In step S48d, when the absolute value of the time integral value IVb when the actual advance value VCTr is displaced toward the advance side is larger than the absolute value of the reference integral value IDb0 and when the actual advance value VCTr is displaced toward the retard side. It is determined whether or not at least one of the conditions that the absolute value of the time integral value IDa is larger than the absolute value of the reference integral value IDa0 is satisfied. Here, the reference integral values IDb0 and IDa0 are integral values of the actual advance value VCTr in the highest response characteristic of the variable valve timing device.

  The absolute values of the integrated values IDb and IDa have a phase difference with the energy input when the actual advance value VCTr is controlled to the target advance value VCTa. For this reason, it is considered that these absolute values increase as the responsiveness decreases. For this reason, when an affirmative determination is made in step S48d, it is determined that there is a deviation from the reference response characteristic, and the process proceeds to step S50.

  The reference integrated values IDa0 and IDb0 are variably set according to the voltage VB of the battery B. That is, these reference integral values IDa0 and IDb0 are values at the current voltage VB. This is done in view of the fact that when the voltage VB of the battery B changes, the duty value D is corrected by the correction coefficient K in the process of step S18 of FIG. For this reason, the reference integrated values IDa0 and IDb0 are variably set in accordance with the voltage VB in order to learn the shift amount of the response characteristic between individual differences excluding the influence of the fluctuation of the voltage VB of the battery B.

  According to the present embodiment described above, the following effects can be obtained in addition to the effects according to the first and second embodiments.

  (12) While quantifying the time change of the actual advance value VCTr based on the time integral value of the Duty value D when feedback control of the actual advance value VCTr to the target advance value VCTa defined by the test pattern Learned. Here, when the feedback control is performed, the duty value D is not constant. Therefore, when the time change of the detected value of the actual advance value VCTr is used, it is associated with which value of the duty value D it corresponds. There is a concern that it is difficult to associate the response characteristic with the deviation amount of the response characteristic. In this regard, in the above-described embodiment, by quantifying the temporal change of the actual advance value VCTr by the time integral value of the duty value D, it is possible to appropriately associate the response characteristic with the shift amount.

(Sixth embodiment)
Hereinafter, the sixth embodiment will be described with reference to the drawings with a focus on differences from the fourth embodiment.

  FIG. 16 shows the procedure of the learning process of the response characteristic deviation amount according to the present embodiment. This process is repeated by the ECU 40, for example, at a predetermined cycle. In FIG. 16, the same steps as those shown in FIG. 14 are given the same step numbers for the sake of convenience.

  In this series of processing, in step S60, it is determined whether or not the target advance value VCTa has changed from the state in which the actual advance value VCTr has been held, and a request to follow the actual advance value VCTr has occurred. To do. If an affirmative determination is made in step S60, it is determined in step S40e whether a learning execution condition is satisfied. Here, as the learning execution condition, in addition to the condition in the fourth embodiment, the absolute value of the change amount ΔD of the duty value D for changing the actual advance value VCTr by the feedback control from the holding state is predetermined. The condition that the value is equal to or less than ε is used. This condition is for determining whether or not the current change in the duty value is a change in a region where the change in the change rate of the actual advance value VCTr with respect to the change in the duty value D is significant. When this condition is satisfied, it is considered that the time change of the actual advance angle value VCTr this time has a strong correlation with the response characteristic due to the shift in the size of the holding dead zone region. For this reason, in order to perform learning only when these learning conditions are satisfied, the process proceeds to step S46c of FIG. 14 when an affirmative determination is made in step S40e.

(Other embodiments)
Each of the above embodiments may be modified as follows.

  -You may change 2nd, 3rd, 5th embodiment by the change with respect to the said 4th Embodiment of the said 6th Embodiment.

  In the fourth embodiment, an integrated value of the deviation Δ may be used instead of the change rate of the deviation Δ.

  According to the change speeds ΔVb and ΔVa in the fourth embodiment, since the change in the target advance value VCTa is included, the target advance value VCTa can be set even when the setting is not one. It can be considered that the deviation amount of the response characteristic can be appropriately reflected regardless of the setting of the advance value VCTa. For this reason, it is particularly effective when applied to a setting in which the test pattern is not uniform.

  In the fifth embodiment, instead of integrating the duty value D, learning may be performed using a change rate toward the holding dead zone after the duty value D has once greatly changed.

  The compensation mode of the variation amount of the voltage VB when controlling the actual advance value VCTr to the target advance value VCTa is not limited to that illustrated in FIG. For example, the correction may be performed by correcting only one or two of the proportional term FBP, the differential term FBD, and the correction amount OFD based on the voltage VB. Further, the same feedback control may be performed regardless of the voltage VB without using the correction coefficient K. In this case, by making the reference value at the time of learning constant regardless of the voltage VB, the learning value OFDa / OFDb includes the deviation amount of the response characteristic due to the fluctuation of the voltage VB. For this reason, when the actual advance value VCTr is controlled to the target advance value VCTa, the fluctuation amount of the voltage VB can be compensated by the learned value OFDa / OFDb.

  The method for learning the deviation amount of the response characteristic limited to the condition where the degree of separation of the operation signal (Duty value D) from the holding point (holding duty value KD) is equal to or less than the specified value is exemplified in the above embodiments. Not limited to things. For example, learning may be performed in a situation where the actual operation signal is fixed at a value at which the degree of separation from the holding point is equal to or less than a specified value by feedback control.

  As a means for quantifying the change over time of the valve characteristic while removing the influence of minute fluctuations in the detected value of the valve characteristic (actual advance value VCTr), an integral value of the actual advance value VCTr and deviation Δ is used. Not exclusively. For example, after a relaxation process such as a filter process for relaxing the change in the actual advance value VCTr is performed, the change speed may be calculated or the change speed of the deviation Δ may be calculated.

  The conditions for determining that the temperature of the hydraulic oil is in a thermal equilibrium state with the surroundings are not limited to those exemplified in the above embodiment. For example, a condition that the cooling water temperature is substantially equal to the outside air temperature may be used.

  -Furthermore, as a learning execution condition, you may remove | exclude the conditions with the case where it is judged that the temperature of hydraulic fluid has implement | achieved the thermal equilibrium state with the circumference | surroundings. In this case, it is desirable to provide dedicated detection means for directly detecting the temperature of the hydraulic oil in the OCV 30 or the variable valve timing mechanism 20.

  -You may remove | exclude the conditions about rotational speed as learning execution conditions. If the discharge pressure can be constant regardless of the rotation speed of the internal combustion engine instead of the oil pump P being driven by the engine, the learning value OFDb based on the rotation speed is calculated when calculating the correction amount OFD. The correction amount OFD can be calculated with high accuracy without correcting / OFDa.

  In the above-described embodiment, the learning of the deviation amount of the response characteristic is performed only when the temperature of the hydraulic oil (cooling water temperature) becomes the specified temperature THW0, but is not limited thereto. For example, learning may be performed separately for each of a plurality of temperature regions obtained by dividing a temperature region that can be taken as the temperature of the hydraulic oil. In this case, it is desirable to set the test pattern separately for each temperature region. In particular, here, it is desirable that the degree of separation of the duty value D with respect to the holding point (holding duty value D) increases as the temperature of the hydraulic oil decreases.

  As a method for setting the duty value D so as to control the actual advance value VCTr to the target advance value VCTa so as to compensate the deviation amount of the response characteristic to be learned, the above-described retained duty value KD is corrected. Not limited to. For example, the relationship between the deviation Δ and the proportional term FBP or the differential term FBD may be corrected.

  The learning method of the retained duty value KD is not limited to the one exemplified in the above embodiment. In short, it is only necessary to provide means for learning while sequentially updating the value of the operation signal when the valve characteristic does not change.

  Instead of adopting the highest response characteristic among the mass produced valve characteristic variable devices, for example, an average response characteristic (center characteristic) may be adopted as a reference characteristic when controlling the valve characteristic.

  The variable valve timing mechanism is not limited to that shown in FIG. For example, a rotating body that rotates in conjunction with the crankshaft 10 may be included in a rotating body that rotates in conjunction with the camshaft 14.

  -The valve characteristic variable device is not limited to the variable valve timing device. For example, as seen in Japanese Patent Application Laid-Open No. 2001-254639, the lift amount of the valve may be variable by displacing the cam shaft in the axial direction by hydraulic oil adjusted by an oil control valve. . Even in such a case, if the valve characteristic of the engine valve is controlled by adjusting the state (pressure) of the working fluid acting on the valve characteristic variable mechanism by the operation signal for the oil control valve, the response of the valve characteristic It is effective to learn the amount of gender deviation.

  The internal combustion engine is not limited to a spark ignition internal combustion engine such as a gasoline engine, but may be a compression ignition internal combustion engine such as a diesel engine.

  DESCRIPTION OF SYMBOLS 10 ... Crankshaft, 14 ... Cam shaft, 20 ... Variable valve timing mechanism (one embodiment of valve characteristic variable mechanism), 30 ... Oil control valve (one embodiment of working fluid adjusting means), 40 ... ECU (valve characteristic control) One embodiment of the device).

Claims (18)

A valve characteristic variable device comprising: a fluid drive type valve characteristic variable mechanism that varies a valve characteristic of an engine valve; and a working fluid adjustment unit that adjusts a state of a working fluid that acts on the valve characteristic variable mechanism. In a valve characteristic control device for an internal combustion engine that is applied and controls the valve characteristic of the engine valve by operating the working fluid adjusting means,
Means for obtaining a detection value of a detection means for detecting a valve characteristic of the engine valve;
Learning means for learning a deviation amount of the response characteristic of the valve characteristic variable device based on a time change of the valve characteristic, using the detection value of the detection means as an input, the learning means being requested from the operating state of the internal combustion engine A valve characteristic control device that performs learning by forcibly changing an operation signal of the working fluid adjusting means for the learning independently of a valve characteristic to be performed. The learning means learns the deviation amount of the response characteristic based on a deviation between a change speed of the valve characteristic quantified with a detection value of the detection means as an input and a reference value. Valve characteristic control device.   The learning is the response characteristic due to a shift of a boundary of a holding dead zone region, which is a region from a holding point at which the valve characteristic of the engine valve is held to a point at which a change rate of the valve characteristic changes rapidly with respect to a change in the operation signal. 3. The valve characteristic control device according to claim 1, wherein the amount of deviation is quantified. The learning means includes the valve for a holding dead zone region in which the operation signal is a region from a holding point at which the valve characteristic of the engine valve is held to a point at which a change rate of the valve characteristic changes rapidly with respect to a change in the operation signal. according to any one of claims 1 to 3, characterized in that performing the learning in a value between the minimum of those boundaries estimated from the response characteristic of the characteristic control device to the boundary of the largest of Valve characteristic control device. Said learning means, in the internal combustion engine temperature lower than the temperature range where the temperature of the working fluid with the operation to converge the, according to any one of claims 1 to 4, characterized in that said learning Valve characteristic control device. Said learning means, any one of claims 1 to 5, characterized in that performing the learning when feedback control of the actual valve characteristic to the target value in accordance with a change in the target value of the valve characteristic The valve characteristic control device described in 1. The operation signal setting means for setting the operation signal of the working fluid adjusting means is further provided so as to control the valve characteristic of the engine valve so as to compensate the deviation amount learned by the learning means. Item 7. The valve characteristic control device according to any one of Items 1 to 6 . 8. The valve characteristic control device according to claim 7, wherein the operation signal setting means takes into account the temperature of the working fluid when setting the operation signal to compensate for the learned shift amount. The operation signal setting means, upon setting the operation signal to compensate for the deviation amount of the learning, claim 7, characterized in that taking into account the pressure of the working fluid output from the working fluid control means or 8. The valve characteristic control device according to 8 . The working fluid adjustment means includes an engine driven pump,
The valve characteristic control device according to claim 9, wherein the operation signal setting means grasps the pressure of the working fluid based on a detected value of the engine rotational speed. Said learning means, any one of claims 1 to 1 0, wherein the working fluid and performs the learning when it is determined that to achieve thermal equilibrium between the ambient The valve characteristic control device described in 1. The valve characteristic is quantified by a one-dimensional parameter,
The working fluid adjusting means displaces the one-dimensional parameter in one direction or the other direction depending on whether the operation signal is on either side of the holding point where the valve characteristic is held ,
It said learning means, the valve characteristic control apparatus according to any one of claims 1 to 1 1, characterized in that said learning separately for each of both sides of the holding points. The valve characteristic varying mechanism according to claim 1 to 1 2, characterized in that the valve timing of the engine valve variable by varying the relative rotational angle difference between the cam shaft to the output shaft of the internal combustion engine The valve characteristic control device according to any one of the above. The learning unit performs the learning based on at least one of a change rate of the valve characteristic based on a detection value of the detection unit and a time integral value of a value obtained by quantifying the valve characteristic. 14. The valve characteristic control device according to any one of 1 to 3 . Said learning means, any claim 1 to 1 4, characterized in that to perform the learning while quantifying the time change of the valve characteristics on the basis of the difference between the target value of the detection value and the valve characteristic of the detection means The valve characteristic control device according to claim 1. The learning means performs the learning when feedback-controlling the detection value of the detection means to the target value according to a change in the target value of the valve characteristic, and the operation used for the feedback control The valve characteristic control device according to any one of claims 1 to 15 , wherein the learning is performed while quantifying a time change of the valve characteristic based on a time integral value of a signal. The operation signal is a duty signal,
The valve characteristic control device according to claim 16 , wherein the learning means takes into account a voltage value of a power supply means for generating the operation signal when learning the deviation amount. The valve characteristic control device according to any one of claims 1 to 17 ,
A valve characteristic control system comprising the variable valve characteristic device.

Images