JP2004360495A - Ignition timing controller of internal combustion engine - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide the ignition timing controller of an internal combustion engine capable of completing the learning of a KCS correction value at an early stage. <P>SOLUTION: The controller for controlling an ignition timing using a basic ignition timing and the KCS correction value is provided with operation condition detecting means 11, 14, 17, 20, 21, 23, 25 and 26 for detecting several kinds of operating conditions of an internal combustion engine, a knocking detecting means 15 and 24 for detecting whether or not the knocking arises, and KCS learning means 15 for learning the KCS correction value using the knocking detecting means 15 and 24. The KCS learning means 15 learns based on the KCS correction value, after specifying a corrected and adjusted amount corresponding to the operation condition detected with the operation condition detecting means 11, 14, 17, 20, 21, 23, 25 and 26 added to the previous KCS corrected value as a new KCS correction value. <P>COPYRIGHT: (C)2005,JPO&NCIPI

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an ignition timing control device for an internal combustion engine that controls the ignition timing for controlling the internal combustion engine.
[0002]
[Prior art]
In an internal combustion engine, ignition timing control is performed in order to obtain the output obtained by combustion with maximum efficiency and to improve exhaust gas purification performance and fuel consumption performance. Here, it is known that in order to obtain the energy generated by combustion as the output most efficiently, it is preferable that the pressure peak inside the combustion chamber is generated at a position slightly delayed from the compression top dead center. For this reason, the ignition timing is determined so that a pressure peak occurs slightly later than the compression top dead center. However, if the ignition timing is too early (too advanced), knocking will occur.
[0003]
The ignition timing at which the internal combustion engine generates the maximum torque is called MBT (Minimum Spark Advance for Best Torque), and depending on the type and rotation speed of the internal combustion engine, the MBT is located near the ignition timing at which knocking starts to occur. is there. Therefore, knocking control (control) is performed so as to obtain an optimal output while suppressing knocking. The knock control repeats gradually advancing when knocking does not occur, gradually retarding when knocking is detected until knocking does not occur, and gradually advancing again when knocking does not occur. ing. The limit ignition timing at which knocking does not occur is referred to as knock limit ignition timing.
[0004]
In normal ignition timing control, the ignition timing is controlled using a basic ignition timing predetermined according to the operating state and a KCS correction value that is a correction amount from the basic ignition timing to the knock limit ignition timing. . That is, control is performed as follows: ignition timing = basic ignition timing + KCS correction value. In the ignition timing control, a correction value other than the KCS correction value may be used together. A knock control device that performs such control is known from, for example, Patent Document 1.
[0005]
[Patent Document 1]
JP 2000-73820 A
[Problems to be solved by the invention]
Since the above-described knock limit ignition timing changes under the influence of the operation state of the internal combustion engine and the weather (external temperature, humidity, etc.) at that time, the KCS correction value is always learned and updated by knock control. The KCS correction value is mapped, and the KCS correction value corresponding to the operating state at that time is taken out and learning is started. However, the KCS correction value at this time was learned in the previous driving condition and weather condition, and the learning was started again based on the KCS correction value. Until the learning is completed, an optimal output is not obtained, which leads to deterioration of fuel efficiency and emission performance. Therefore, further improvement has been desired.
[0007]
Therefore, an object of the present invention is to provide an ignition timing control device for an internal combustion engine that can complete learning of a correction value (KCS correction value) in knock control earlier.
[0008]
[Means for Solving the Problems]
An ignition timing control device for an internal combustion engine according to claim 1, wherein a basic ignition timing predetermined according to an operation state of the internal combustion engine and a KCS correction value which is a correction amount from the basic ignition timing to a knock limit ignition timing. And the ignition timing is controlled by using an operating state detecting means for detecting various operating states of the internal combustion engine, a knocking detecting means for detecting whether knocking has occurred, and a knocking detecting means. KCS learning means for learning a KCS correction value. The KCS learning means described above calculates a new KCS correction value by adding a correction correction amount corresponding to the driving state detected by the driving state detection means to the previous KCS correction value, and then sets the KCS correction value Learning is performed based on.
[0009]
According to a second aspect of the present invention, in the first aspect, the operating state detecting means detects at least one of an engine speed, an engine load, a valve timing, an intake temperature, and an intake humidity as an operating state quantity. .
[0010]
According to a third aspect of the present invention, in the first or second aspect of the present invention, the KCS correction value is mapped on the basis of the operating state, and the KCS learning value learned by the KCS learning means is replaced with another KCS learning value. There is further provided learning reflecting means for reflecting the KCS learning value in the operating region.

[0011]
BEST MODE FOR CARRYING OUT THE INVENTION
An embodiment of the control device of the present invention will be described below. FIG. 1 shows an engine 1 having a control device according to the present embodiment.
[0012]
The engine 1 described in the present embodiment is a multi-cylinder engine, but here only one cylinder is shown in FIG. 1 as a cross-sectional view. In the engine 1, outside air is taken in as intake air through an intake passage 2, and the intake air is mixed with fuel injected from an injector 4 immediately before a cylinder 3 to form an air-fuel mixture. The air-fuel mixture is sucked into the cylinder 3, compressed by the piston 5, ignited by the ignition plug 6, and burned. At this time, the pressure in the cylinder rises due to the combustion, and this is taken out as an output through the piston 4 and the connecting rod. The interior of the cylinder 3 and the intake passage 2 are opened and closed by an intake valve 7. The exhaust gas after combustion is exhausted to the exhaust passage 8. The interior of the cylinder 3 and the exhaust passage 8 are opened and closed by an exhaust valve 9.
[0013]
On the intake passage 2, an air cleaner 10, an air flow meter 11, a throttle valve 12, and the like are arranged from the upstream side. The air cleaner 10 is a filter that removes dust and dirt from the intake air. The air flow meter 11 of the present embodiment is of a hot wire type, and detects an intake air amount as a mass flow rate. The air flow meter 11 also has a built-in thermistor-type intake air temperature sensor, and the air flow meter 11 also has a function as an intake air temperature sensor. Downstream of the air flow meter 11, a throttle valve 12 for adjusting the amount of intake air is disposed.
[0014]
The throttle valve 12 of the present embodiment is a so-called electronically controlled throttle valve. An operation amount of the accelerator pedal 13 is detected by an accelerator positioning sensor 14, and an electronic control unit (ECU) 15 determines an opening degree of the throttle valve 12 based on the detection result and other information amount. The throttle valve 12 is opened and closed by a throttle motor 16 provided in association with the throttle valve. A throttle positioning sensor 17 for detecting the opening of the throttle valve 12 is also provided in association with the throttle valve 12. The ECU 15 includes a CPU, a ROM, a RAM, and the like. Further, a surge tank is disposed downstream of the throttle valve 12, and a humidity sensor 25 is mounted in the surge tank.
[0015]
Further, the above-mentioned ignition plug 6 is connected to the ECU 15 via an ignition coil 18 and an igniter 19. A crank positioning sensor 20 for detecting an engine speed and a piston position is mounted near the crankshaft of the engine 1. An intake valve 7 (and an exhaust valve 9) is mounted near a camshaft on the intake side. The cam positioning sensor 21 for detecting the opening / closing timing of (2) is attached. The ECU 15 determines the ignition timing based on the detection results of the crank positioning sensor 20 and the cam positioning sensor 21. In the ignition plug 6, the igniter 19 operates as a switch based on an ignition signal from the ECU 15, and the ignition coil 18 generates a high voltage for ignition, whereby the ignition is performed.
[0016]
The ECU 15 is connected to the igniter 19, the crank positioning sensor 20, the cam positioning sensor 21, the above-described air flow meter 11, the accelerator positioning sensor 14, the throttle motor 16, the throttle positioning sensor 17, and the like. These sensors and actuators send detection results to the ECU 15 or are controlled by signals from the ECU 15. Although not shown, the ECU 15 is also connected to an air-fuel ratio sensor and the like used for air-fuel ratio control for effectively purifying the exhaust gas by the exhaust gas purification catalyst 22 disposed on the exhaust passage 8. It is also responsible for control and air-fuel ratio control.
[0017]
The ECU 15 is also connected with a water temperature sensor 23 and a knock sensor 24 for detecting the temperature of the engine cooling water. The ECU 15 also performs knock control (control) based on the detection result of the knock sensor 24. Knock sensor 24 is fixed to the engine block, and detects engine vibration. The ECU 15 determines whether or not knocking has occurred in the engine 1 based on the detection result to determine whether or not knocking-specific vibration has occurred. Further, an external air temperature sensor 26 for detecting an external air temperature is also connected to the ECU 15.
[0018]
Further, although not shown, the engine 1 of the present embodiment has a continuously variable valve timing variable mechanism on the intake valve 7 side. The variable mechanism in the present embodiment is a known mechanism, and is built in a sprocket of a camshaft on the intake valve 7 side. This variable valve timing mechanism changes the rotational phase between a sprocket and a camshaft by hydraulic pressure generated by an oil pump. The valve between the oil pump and the sprocket is controlled by the ECU 15, and the above-described oil pressure is controlled by the valve control to control the opening / closing timing of the intake valve 7.
[0019]
Hereinafter, the ignition timing control in the present embodiment will be described. The ignition timing is determined by (basic ignition timing + KCS correction value [+ other correction values as necessary]) as described above. The basic ignition timing is provided with a certain margin with respect to the ignition timing at which knocking occurs in consideration of fluctuations in climatic conditions and the like. For this reason, up to MBT (optimal ignition timing) near knocking, the ignition timing is reduced (advanced) using the KCS correction value so that an optimum output is obtained.
[0020]
As described above, the knock limit ignition timing (= basic ignition timing + KCS correction value) differs depending on the individual difference between the engines, and also differs depending on the operating state of the engine, the aging state, the climatic condition at that time, and the like. For this reason, the KCS correction value is constantly learned by the knock control system. However, conventionally, a KCS correction value corresponding to the operating state is extracted from the KCS correction value map, and learning is performed on the extracted KCS correction value. For example, a two-dimensional map (or a multi-dimensional map) of the engine speed and the engine load is made for the KCS correction value, and the map is made based on the operating state (engine speed and engine load). Learning was started by extracting the corresponding KCS correction value.

[0021]
However, the KCS correction value taken out at this time is updated under the driving state and the weather condition at the time of the previous learning, and the KCS correction value is re-learned on the basis of the driving state and the weather state despite the change. It took time to learn to do. Therefore, in the present embodiment, the extracted KCS learning value is temporarily corrected according to various conditions affecting the knock limit ignition timing, and learning is started using the corrected KCS correction value. By performing the correction once, learning of the KCS correction value can be completed earlier.
[0022]
Further, in the present embodiment, the KCS learning value for which learning has been completed is reflected in an adjacent driving region on the map. At this time, weighting is applied to an adjacent driving area on the map and then reflected. It can be said that the operating region adjacent to the operating region (operating state) in which the learning has been completed has a high probability that re-learning will be started. Therefore, by reflecting the already completed learning value in the adjacent operation region, it is possible to improve the accuracy of the initial KCS correction value when the re-learning is started and complete the re-learning earlier. it can.
[0023]
The above control will be described based on the flowchart shown in FIG. First, the operating state of the engine 1 is detected by the various sensors described above (step 200). The state quantities to be detected include the engine speed NE detected by the crank positioning sensor 20, the engine load KL obtained from the intake air amount detected by the air flow meter 11 and the throttle opening detected by the throttle positioning sensor 17, The valve overlap amount VT obtained by the positioning sensor 20 and the cam positioning sensor 21, the intake air temperature T _in [° C.] detected by the intake air temperature sensor incorporated in the air flow meter 11, and the intake humidity AH detected by the humidity sensor 25 [ g / kg], the atmospheric pressure Pa [kPa] detected by the atmospheric pressure sensor 26, and the like.
[0024]
The humidity used here is an absolute mass expressed as (mass of water vapor) / (mass of dry air) in a certain humid air, and the unit is [g / kg]. The unit may be expressed as [g / kg '] or [g / kgDA], and [kg'] or [kgDA] indicates the mass of dry air (Dry Air). Regarding the KCS learning value described below, the retard side is treated as positive.
[0025]
After step 200, the basic ignition timing and the KCS learning value (correction value) KCS _N-1 corresponding to the state quantity detected in step 200 are read from each map (step 205). Further, based on the state quantity detected in step 200, it is determined whether or not the operation state has changed between the time of the previous reading and the current time (step 210). Whether or not the operating state has changed is determined by calculating the difference between the previous value and the current value for each information amount, and determining whether the difference exceeds a predetermined threshold. As a result, if it is determined that there is no state change, the control of the flowchart in FIG. 2 is finished, and the relearning of the KCS learning value is not performed (the current KCS learning value is maintained as it is).
[0026]
On the other hand, if step 210 is denied, that is, if it is determined that a change in the operating state has occurred between the time of the previous reading and this time, the changed state quantity is specified (step 215), and the specified change is determined. A KCS correction correction amount ΔKCS _N−1 is calculated based on the state quantity (step 220). This calculation is performed as a function f of ΔST ₁ , ΔST ₂ , and ΔST ₃ as shown in the following equation (1).
ΔKCS _N−1 = f (ΔST ₁ , ΔST ₂ , ΔST ₃ ) (1)
[0027]
Here, ΔST ₁ is a value focusing on a change in the atmospheric pressure as a change in the operating state of the engine 1 as shown by the following equation (2), and includes an engine speed NE, an engine load KL, and a valve overlap amount VT. , obtained as a function _{g 1} of the difference of the atmospheric pressure _{(Pa-P 0)} with respect to the reference state.
ΔST ₁ = g ₁ (NE, KL, VT, (Pa−P ₀ )) (2)
NE, KL, VT, and Pa in the equation (2) are as described above, and P ₀ [kPa] is the atmospheric pressure in the reference state.
[0028]
ΔST ₂ is a value focusing on a change in the intake air humidity as a change in the operating state of the engine 1 as shown by the following equation (3), and represents the engine speed NE, the engine load KL, and the intake air humidity relative to the reference state. obtained as a function _{g 2} of the difference (AH-AH _0).
ΔST ₂ = g ₂ (NE, KL, (AH-AH ₀ )) (3)
NE, KL, and AH in the equation (3) are as described above, and AH ₀ [g / kg] is the humidity in the reference state.
[0029]
Further, ΔST ₃ is a value focusing on a change in the intake air temperature as a change in the operating state of the engine 1 as shown by the following equation (4), and indicates the engine speed NE, the engine load KL, and the intake air temperature with respect to the reference state. Obtained as a function of the difference (T _in -T ₀ ).
ΔST ₃ = g ₃ (NE, KL, (T _in −T ₀ )) (4)
NE, KL, and AH in the expression (4) are as described above, and T ₀ [° C.] is the temperature in the reference state.
[0030]
A value obtained by adding ΔKCS _N−1 calculated in this way to the KCS learning value KCS _N−1 read in step 205 is defined as KCS ′ _N−1 (= KCS _N−1 + ΔKCS _N−1 ), and this is newly defined. (Step 225), and re-learning is performed in subsequent steps. In the past, since KCS _N-1 was re-learned as it was, it took time to complete the learning. Here, ΔKCS _N−1 is calculated based on the state change amount, and the result of adding ΔKCS _N−1 to KCS _N−1 is treated as a new KCS learning value. By starting the re-learning after the reflection, the re-learning can be completed early.
[0031]
From the beginning, it is conceivable to map the KCS learning value using many parameters. If the case described above in Examples, NE, KL, VT, Pa , AH, it is conceivable to create a map using all parameters of _{T in.} However, as the number of parameters increases, the calculation load increases. Further, even if the dimension of the parameter is increased, a change in the state of the engine itself or a change over time occurs, so that learning is indispensable. If the number of areas in the map increases, the chance of re-learning for each area decreases, so increasing the number of parameters may not always lead to improving the accuracy of the map itself. Furthermore, a huge number of man-hours are required to generate the map. By using ΔKCS _N−1 as in the present embodiment, the KCS learning value can be updated more accurately while realizing the early completion of learning without complicating the map.
[0032]
The ignition timing (= basic ignition timing + KCS'N _-1 ) is set using KCS'N _-1 set as the new KCS learning value in step 225 (step 230), and the knock control system is set using this ignition timing. Is performed again (step 235). Assuming that the correction amount for the re-learning is KCS " _N-1 , the ignition timing after the re-learning is represented by (basic ignition timing + KCS'N _-1 + KCS" _N-1 ) (step 240). . Here, (KCS ′ _N−1 + KCS ″ _N−1 ) is set as a final KCS learning value KCS _N (step 245).
[0033]
When the final KCS learning value KCS _N is determined, the final KCS learning value KCS _N is reflected after weighting the adjacent area on the map (step 250). FIG. 3 schematically shows this. FIG. 3 shows a map of the KCS learning value, which is formed as a two-dimensional map including the engine speed NE and the engine load KL. In FIG. 3, it is assumed that the operating region is partitioned and mapped as squares, and the final KCS learning value KCS _N for the operating region X indicated by hatching is determined. This is reflected after weighting the eight adjacent operation regions.
[0034]
A known method can be used as a weighting method. For example, to store the average value of the KCS learning value in the operation region where the current operating region X learning adjacent to the final KCS learning value KCS _N completing already been stored as a new KCS learning value of the operating region. Since the operating region adjacent to the operating region X of the engine 1 that has been learned this time has a high probability of re-learning, the latest learning result on the operating region X for which re-learning has been completed is reflected in this manner, so that re-learning is performed. When, the re-learning can be completed earlier, and more accurate learning can be performed.
[0035]
In the above-described control, sensors for detecting various operating state quantities serve as operating state detecting means, knock sensor 23 and ECU 15 serve as knocking detecting means, an ignition system including ECU 15 serves as KCS learning means, and ECU 15 serves as a KCS learning means. It functions as learning reflection means.
[0036]
The present invention is not limited to the embodiments described above. For example, in the above-described embodiment, the KCS correction value for which learning has been completed is reflected only in an adjacent driving region, but may be reflected further in a region outside the driving region or the whole. Further, in the above embodiment, the weight is reflected after being applied, but this does not necessarily mean that the weight must be applied. For example, the same KCS correction value may be reflected only in an adjacent area, or the same KCS correction value may be reflected in an adjacent area, and weight may be applied to an area outside the KCS correction value. .
[0037]
Further, in the above-described embodiment, the KCS correction value is re-learned (steps 225 to 235 in FIG. 2) regardless of the value of ΔKCS calculated by equation (1). If ΔKCS is smaller than the predetermined value, steps 225 and 230 may be omitted and relearning may be performed. That is, when ΔKCS is small, re-learning is started without reflecting ΔKCS, whereby the calculation load is reduced, and re-learning can be completed earlier.
[0038]
Furthermore, in the above-described embodiment, the operation state quantity is almost directly detected by sensors. However, when detecting the operation state quantity as the operation state detection means, detection using a so-called model for estimating and estimating a physical phenomenon by calculation using a mathematical method may be performed.
[0039]
【The invention's effect】
According to the ignition timing control device for an internal combustion engine according to the first aspect, the KCS learning means adds the correction correction amount corresponding to the operating state detected by the operating state detecting means to the previous KCS correction value. After setting a new KCS correction value, learning is performed based on the KCS correction value. Therefore, the learning can be completed earlier by the amount of the correction correction than when the learning is started from the previous KCS learning value. Since this correction correction amount is determined based on the driving state at that time, there is an advantage that not only the completion of learning can be accelerated but also the learning accuracy can be improved.
[0040]
According to the third aspect of the present invention, the KCS learning value learned by the KCS learning means is reflected on the KCS learning value of another operating region, so that another operating region is learned. Since the new learning value is already reflected in the learning time, the learning time can be reduced more effectively, and the learning accuracy can be further improved.
[Brief description of the drawings]
FIG. 1 is a configuration diagram showing a configuration of an internal combustion engine having one embodiment of an ignition control device of the present invention.
FIG. 2 is a flowchart of control by an embodiment of the ignition control device of the present invention.
FIG. 3 is an explanatory diagram schematically showing a map of KCS learning values.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Engine, 11 ... Air flow meter (built-in intake air temperature sensor: operating state detecting means), 12 ... Throttle valve, 13 ... Accelerator pedal, 14 ... Accelerator positioning sensor (operating state detecting means), 15 ... ECU (Knock detecting means. KCS learning means / learning reflection means), 17: throttle positioning sensor (operating state detecting means), 18: ignition coil, 19: igniter, 20: crank positioning sensor (operating state detecting means), 21: cam positioning sensor (operating state) Detecting means), 22: exhaust purification catalyst, 23: water temperature sensor (operating state detecting means), 24: knock sensor (knocking detecting means), 25 ... humidity sensor (operating state detecting means), 26: atmospheric pressure sensor (operating state) Detection means).

Claims (3)

Ignition timing control for an internal combustion engine that controls ignition timing using a basic ignition timing predetermined according to the operating state of the internal combustion engine and a KCS correction value that is a correction amount from the basic ignition timing to the knock limit ignition timing In the device,
Operating state detecting means for detecting various operating states of the internal combustion engine,
Knocking detection means for detecting whether knocking has occurred,
KCS learning means for learning the KCS correction value using the knocking detection means,
The KCS learning means sets a new KCS correction value by adding a correction correction amount corresponding to the driving state detected by the driving state detection means to the previous KCS correction value, and then, based on the KCS correction value. Ignition timing control means for an internal combustion engine. 2. The internal combustion engine according to claim 1, wherein the operating state detecting unit detects at least one of an engine speed, an engine load, a valve timing, an intake air temperature, and an intake air humidity as an operating state quantity. 3. Ignition timing control device. The KCS correction value is mapped based on the operating state, and further includes learning reflecting means for reflecting the KCS learning value learned by the KCS learning means to the KCS learning value in another operation region. The ignition timing control device for an internal combustion engine according to claim 1 or 2, wherein:

Images