JPS6232259A - Monitor of learning controller for internal combustion engine - Google Patents

Abstract

PURPOSE:To keep off the worsening of engine driving performance due to a breakdown of learning contents, by interpolating a learning compensation value to the initial value at a time when the learning compensation value stored in a memory device is unusual, and constituting the learning so as to be performed from the initial stage. CONSTITUTION:A learning controller is provided with a fundamental control value setting device B, setting a fundamental control value corresponding to the control desired value of an engine A, and a memory device C storing a learning compensation value compensating the fundamental control value. Also, it is provided with a setting device E, setting a feedback compensation value for compensating the fundamental control value according to a deviation between the actual control value and the control desired value, and a renewal device F setting a new learning compensation value from the compensation value and the learning compensation value searched out of the memory device C by a searching device D and renewing it. In this case, there is also provided with a judging device J which judges whether the learning compensation value read out of the memory device C is normal or not by a reading device I, and when anything unusual is judged, the said learning compensation value is made so as to be interpolated to the initial value by an initiating device K.

Description

[Detailed description of the invention] <Industrial application field> The present invention relates to a monitoring device for a learning control device for an internal combustion engine.

<Prior Art> For example, in an electronically controlled fuel injection type internal combustion engine that controls the air-fuel ratio, components such as an air flow meter, a fuel injection valve, a pressure regulator, etc.

Due to factors such as variations between products (control device) and changes over time, the base air-fuel ratio deviates from λ = 1 in the λ control region. The air-fuel ratio is feedback-controlled.

However, if the base air-fuel ratio, which is the control target value, deviates from λ-1, it takes time to stabilize the step in the base air-fuel ratio to λ=1 by feedback control when the operating range changes significantly.

Therefore, in Japanese Unexamined Patent Publication No. 59-203828, the applicant set the base air-fuel ratio to λ = 1 through learning, thereby eliminating the deviation of the base air-fuel ratio from λ = 1 during transients and improving controllability. We proposed a base air-fuel ratio learning control device that aims to reduce the cost of catalysts by improving exhaust characteristics.

That is, the learning correction coefficient α corresponding to engine operating conditions such as engine speed and load is stored in the RAM. When calculating the injection amount Ti, which is a control value, the basic injection amount 'rp, which is a basic control value, is set as α, as shown in the following equation. Correct with.

T i =Tp XC0EFxα×αo+'['s and α. Proceed with the following steps to learn.

l) Detect the engine operating conditions and α in a steady state.

ii) α that has been learned and stored up to now in correspondence with the engine operating conditions. Search for.

iii) said α and said α. Tokaraα. +Δα/M is newly set to obtain the above α. and update.

In addition, Δα indicates the amount of deviation of α from the reference value (α1), and Δα
α=α−α, and the reference value α1 is generally 1.

Further, M is a constant (greater than 1).

By the way, the learning correction coefficient α is updated and stored in the RAM. In order to maintain the power, the RAM is energized even after the engine is stopped, but if the energization to the RAM is turned off or momentarily interrupted, the contents stored in the RAM will be destroyed. Therefore, as shown in Japanese Patent Laid-Open No. 59-211742, the present applicant has proposed that when the engine is stopped, power to the RAM is turned off.
We have proposed a monitoring device for a learning control device that returns the learning correction coefficient stored in the RAM to its initial value when the learning control device is turned off or momentarily interrupted.

<Problems to be Solved by the Invention> However, in such a conventional monitoring device for a learning control device, when power to the RAM is turned off or momentarily interrupted, the learning correction coefficient stored in the RAM is reset to the initial value. For example, when the RA is in a strong electric field such as a radio wave,
If M is exposed, the memory contents of the RAM will be partially or completely destroyed even if the RAM is normally energized. In particular, if the memory contents of the RAM are partially destroyed, for example, the fuel injection amount changes greatly in a specific operating range of the engine, deteriorating the accuracy of air-fuel ratio control and causing deterioration of driving performance such as exhaust characteristics.

The present invention was made in view of the above circumstances, and R
It is an object of the present invention to provide a monitoring device for a learning control device that can determine whether a learning correction amount stored in a RAM is normal regardless of whether the AM is energized or not.

<Means for Solving the Problems> Therefore, as shown in FIG. A storage means C for storing learning correction amounts set in correspondence and correcting the basic control value, and a learning correction amount retrieval means for retrieving the learning correction amount from the storage means C under the same operating conditions as the actual control value. a feedback correction amount setting means E for comparing an actual control value and the control target value and setting a feedback correction amount for correcting the basic control value so as to bring the actual control value closer to the control target value; A learning correction amount update that sets a new learning correction amount from the correction amount and the retrieved learning correction amount and updates the learning correction amount stored in the storage means C under the same operating conditions to this new learning correction amount. means F, control value calculation means G for calculating a control value based on the basic control value, learning correction amount and feedback correction amount, and control means H for controlling the control means of the engine A based on the calculated control value. and a successive means I for reading out the learning correction amount from the storage means C, a determining means J for determining whether the read learning correction amount is a normal value or an abnormal value, and a determining means J for determining whether the read learning correction amount is a normal value or an abnormal value. A rewriting means for rewriting the learning correction amount stored in the storage means C to an initial value is provided.

In this way, when the learning correction amount stored in the storage means is an abnormal value, the learning correction amount is rewritten to the initial value and learning is performed from the beginning.

<Example> An example of the present invention will be described below based on FIGS. 2 to 4. In this embodiment, a learning control device related to air-fuel ratio control will be described.

In FIG. 2, 1 is a CPU and 2 is a P-ROM.

3 is a CMOS-RAM for learning control as a storage means.
, 4 is an address decoder.

Analog input signals to the CPUI for controlling the fuel injection amount include an intake air flow rate signal from the hot wire airflow meter 5, a throttle opening signal from the throttle sensor 6, a water temperature signal from the water temperature sensor 7, and an oxygen sensor 8. Exhaust oxygen concentration signal from.

There are battery voltages from the battery 9, which are adapted to be input via an analog input interface 10 and an A/D converter 11. (2) 2 is an A/D conversion timing controller.

Digital human input signals include ON-OFF signals from the idle switch 13, start switch 14, and neutral switch 15, and these are inputted via the digital input interface 16.

In addition, a reference signal every 180 degrees and a position signal every 1 degree from the crank angle sensor 17 are input via the one-shot multi-circuit 18. Further, a vehicle speed signal from a vehicle speed sensor 19 is inputted via a waveform shaping circuit 20.

The output signal (driving pulse signal to the fuel injection valve) from the CPU 1 is sent to the fuel injection valve 22 via the current waveform control circuit 21.
It is now sent to

Further, the CPU 2 and the P-ROM 2 are connected to a battery 9, an engine key switch (not shown), and a stabilizing power supply circuit (not shown).
(not shown), a power supply voltage is applied thereto. Said CMO
3-A power supply voltage is applied to the RAM 3 via the stabilized power supply circuit when the engine key switch is on, and from a hack-up power supply circuit (not shown) when the engine key switch is off.

Here, the CPU 1 operates according to a program (stored in the ROM 2) based on the flowcharts shown in FIGS. 3 and 4. Here, the CPU includes basic control value setting means, learning correction amount searching means, feedback correction amount setting means, learning correction amount updating means, control value calculation means,
The CPU 2 and the fuel injection valve 22 constitute a control means that also serves as a determination means and a rewriting means.

Next, the flowcharts in FIGS. 3 and 4 will be explained.

First, to explain the learning control of the air-fuel ratio, the basic injection amount Tp (=K
xQ/N).

In S2, various increase correction coefficients C0EF, which will be described later, are set.

In S3, the output from the oxygen sensor 8 and the slice level are compared and the air-fuel ratio feedback correction coefficient α is set by proportional-integral control.

In S4, a voltage correction numerator s is set based on the battery voltage from the battery 9.

In S5, engine speed N and basic injection amount (load) ”rp
Learning correction coefficient α from. Search for. Note that the learning correction coefficient α for the rotation speed N and the basic injection amount 'rp.

The map is stored in rewritable RAM3,
All values are α when learning has not started.

= 1.

86 to S9 are provided to detect a steady state, and S6 determines a change in vehicle speed based on the signal from the vehicle speed sensor 19, and S7 determines the gear position based on the signal from the neutral switch 15. Then, in S8, a change in the throttle opening degree is determined based on the signal from the throttle sensor 6, and in S9, it is determined whether a predetermined time has elapsed, and if it is within the predetermined time, the process returns to S6. In this way, if the change in vehicle speed is less than or equal to a predetermined value within a predetermined time, the gear is engaged, and the change in throttle opening is less than or equal to a predetermined value, it is determined that the vehicle is in a steady state, and slo and so are selected. learning correction coefficient α
. to make corrections. In addition, if the change in vehicle speed exceeds a predetermined value at any point within a predetermined time, if the vehicle becomes neutral, or if the change in throttle opening exceeds a predetermined value, it is determined that the vehicle is in a transient state, Learning correction coefficient α for sio and so.

Do not make any modifications.

In addition, methods such as rich/lean reversal of the oxygen sensor output, α status, and combination of driving parameters can be considered to detect the steady state, but when considering matching with the response, it is possible to detect the change in vehicle speed, gear It is easy to make a judgment based on the condition that a predetermined period of time elapses after the combination of throttle opening changes (positions other than neutral) reaches a predetermined state.

Learning correction coefficient α when the steady state is determined.

The modification will be made as follows.

Learning correction coefficient α retrieved by SIO based on the current air-fuel ratio feedhack correction coefficient α, engine speed N, and basic injection amount Tp. and a new learning correction coefficient α.

seek.

α. −α. +Δα/M; M is a constant.

ΔαTen α−α1 New learning correction coefficient α in Sll. is written to the corresponding engine speed N and basic injection ITp in RAM3. That is, the data in RAMa is updated.

Then, in S12, the injection amount Ti is calculated according to the following equation.

Ti””TpXCOEFXαXαG +TsHere, C
0EF is various correction coefficients, and Ts is a battery correction coefficient.

Next, the monitoring function will be explained according to the flowchart shown in FIG.

When the engine key switch is turned on and power supply voltage is applied from the battery to CPU 1, CMO3-RA is
Learning correction coefficient α stored in M3 based on engine speed N and basic injection amount Tp. is not read for each area of CMO3-RAM3.

In S22, the learning correction coefficient α is read out. The learning correction coefficient α is read out by comparing the learning correction coefficient with the upper limit value, which is the maximum normal value. When the value exceeds the upper limit, the process proceeds to S23, and when it is less than the upper limit, the process proceeds to S24.

In S24, the learning correction coefficient α is read out. and the lower limit value (for example, 1) where the learning correction coefficient is the minimum normal value,
The read learning correction coefficient α. is less than the lower limit value, S
The process proceeds to S23, and when the value is equal to or greater than the lower limit value, the process proceeds to S25.

In this way, the learning correction coefficient α is read out.

When exceeds the range from the upper limit to the lower limit, CM
It is determined that the learning correction coefficient stored in O3-RAM3 has been destroyed due to a failure of the hack-up circuit or exposure of CMO5-RAM3 to a strong electric field such as a radio wave, and is an abnormal value. Then, in S23, all learning correction coefficients stored in the CMO3-RAM3 are rewritten to initial values (=1).

Also, when it is determined that the value is within the range from the upper limit value to the lower limit value, it is determined in S25 whether or not the check of all areas of CMO3-RAM3 has been completed, and if it has not been completed, S26
After changing the region to be determined in step S21, the learning correction coefficient of another region is read out and it is determined whether the coefficient is a normal value or not.

As explained above, it is determined whether the learning correction coefficient stored in CMO3-RAM3 is within the range from the upper limit value to the lower limit value, and if it is outside the range, the learning correction coefficient is determined to be abnormal and the initial value is set. Since the learning correction coefficients for all regions are rewritten, even if CMO3-RAM3 is exposed to strong electric fields such as radio waves and some of the learning contents of CMO3-RAM3 are destroyed, the learning correction coefficients will be rewritten from the initial values in all operating regions of the engine. will be resumed.

Therefore, even if a portion is destroyed, the fuel injection amount can be prevented from changing significantly in a specific operating range, and deterioration of exhaust characteristics can be prevented. In particular, since abnormalities due to destruction are determined from the learning correction coefficients stored in the CMO3-R-AM3, abnormalities can be reliably monitored, thereby improving the reliability of the learning contents of the CMO3-RAM3.

In addition, although the learning control of the air-fuel ratio was described in this embodiment,
It can also be applied to devices that perform learning control on ignition timing or learning control on idle speed.

<Effects of the Invention> As explained above, in the present invention, when the learning correction amount stored in the storage means is abnormal, the learning correction amount is rewritten to the initial value and learning is performed from the beginning. It is possible to reliably monitor abnormalities even when the learning contents are destroyed due to exposure to strong electric fields, thereby improving the reliability of the learning contents while preventing deterioration of engine operating performance due to destruction of the learning contents. can.

[Brief explanation of drawings]

FIG. 1 is a diagram corresponding to claims of the present invention, FIG. 2 is a configuration diagram showing an embodiment of the present invention, FIG. 3 is a flowchart of air-fuel ratio learning control same as above, and FIG. 4 is another flowchart same as above. l...CPU 3...CMO3-RAM8...
・Oxygen sensor 22...Fuel injection valve patent applicant Japan Electronics Co., Ltd. Agent Patent attorney Fujio Sasashima Figure 3 Figure 4

Claims (1)

[Claims] basic control value setting means for setting a basic control value corresponding to a control target value of a controlled object in an engine; storage means for storing a learning correction amount set corresponding to the control value and correcting the basic control value; learning correction amount retrieval means for retrieving a learning correction amount from the storage means under the same operating conditions as the control value; a feedback correction amount setting means for correcting the basic control value, and setting a new learning correction amount from the feedback correction amount and the retrieved learning correction amount, and storing the new learning correction amount under the same operating conditions. learning correction amount updating means for updating the learning correction amount stored in the means; control value calculation means for calculating a control value based on the basic control value, the searched learning correction amount and the feedback correction amount; A learning control device for an internal combustion engine, comprising: a control means for controlling the controlled object based on a value; An internal combustion engine characterized by comprising a determining means for determining whether the value is a normal value or an abnormal value, and a rewriting means for rewriting the learning correction amount stored in the storage means to an initial value when the value is determined to be an abnormal value. Monitoring device for learning control equipment of the engine.