JPS60153448A - Feedback controller with learning function - Google Patents

Abstract

PURPOSE:To improve control accuracy in time of transient operation, by judging a progress degree of learning at every operating region and, on the basis of the judgment, interpolatively calculating a learning value at the operating region being small in the learning progress degree from another learning value at an operating region being large in the learning progress degree in and around the operating region. CONSTITUTION:In the case where an operating region is carried in an arrow direction, when it exists in an unlearned region (a), both upper and lower learning regions A and B are searched for, then a learning compensation factor alpha0 at both these regions A and B is read out, calculating the learning compensation factor alpha0 at the unlearned region by means of proportional interpolation, whereby an injection quantity Ti is calculated. A driving pulse signal commensurate to this injection quantity is given to a fuel injection valve 22 at the specified timing via a current waveform control circuit 21, and such an estimated learning compensation factor as being high in reliability is used even for the unlearned region, thus injection quantity control takes place. With this constitution, a step difference in an air-fuel ratio between these learning and unlearning regions is brought to nothing at all, thus exhaust emission in a transient state is prevented from getting worse.

Description

DETAILED DESCRIPTION OF THE INVENTION Technical Field The present invention relates to a feedback control device with a learning function applied to an electronically controlled fuel injection control device for an internal combustion engine.

(Background Art) In an electronically controlled fuel injection type internal combustion engine, the injection amount Ti is determined by the following equation.

Ti = TpXcolEFxα+Ts Here, Tp is the basic injection amount Tp=KXQ/N It is.

K is a constant, Q is the intake air flow rate, and N is the engine speed.

C0EF is various correction coefficients. α is an air-fuel ratio feedback correction coefficient for air-fuel ratio feedback control (λ control) to be described later. Ts is the voltage correction bone;
This is to correct changes in the injection amount of the electromagnetic fuel injection valve due to fluctuations in battery voltage.

Regarding λ control, an 02 sensor is installed in the exhaust system to detect the actual air-fuel ratio, determine whether the air-fuel ratio is richer or leaner than the stoichiometric air-fuel ratio based on the slice level, and adjust the amount of fuel injected to achieve the stoichiometric air-fuel ratio. Therefore, the air-fuel ratio is maintained at the stoichiometric air-fuel ratio by determining the above-mentioned air-fuel ratio fee t' bank supplementary correction coefficient and changing this α.

Here, the value of the air-fuel ratio feedback correction coefficient α is changed by proportional-integral (PI) control to achieve stable control.

In other words, by comparing the 02 sensor output and the slice level, if the air-fuel ratio is higher or lower than the slice level, the air-fuel ratio will not be suddenly richer or leaner, and if the air-fuel ratio is richer (leaner). First lower (raise) by P,
Then, gradually increase the air-fuel ratio in 1-minute increments to make the air-fuel ratio leaner (
darken).

However, in a region where λ control is not performed, the air-fuel ratio is clamped to α-I, and a desired air-fuel ratio is obtained by setting various correction coefficients COFF.

By the way, if the base air-fuel ratio at α-1 in the λ control region could be set to the stoichiometric air-fuel ratio (λ-1), foot bank control would not be necessary. valve, pressure regulator.

Since the base air-fuel ratio deviates from λ-1 due to factors such as variations in the control unit (control unit), changes over time, non-linearity of the pulse-flow characteristics of the fuel injector, and changes in operating conditions and environment, feedback control is performed. It is carried out.

However, if the base air-fuel ratio deviates from λ-1, it takes time to set the step in the base air-fuel ratio to λ-1 by feedback control when the operating range changes significantly. And for this we need the constant of proportionality and integration (P/T
) is increased, resulting in overshoot or undershoot, resulting in poor controllability. In other words, if the base air-fuel ratio deviates from λ-1, the air-fuel ratio will be controlled within a range that deviates considerably from the stoichiometric air-fuel ratio.

As a result, the three-way catalyst has to be operated at a point where its conversion efficiency is poor, which not only increases costs due to an increase in the amount of precious metals in the catalyst, but also necessitates replacement of the catalyst due to deterioration in conversion efficiency due to deterioration of the catalyst. There was a problem.

Therefore, by setting the base air-fuel ratio to λ-1 through learning, it is possible to eliminate the deviation from λ = 1 caused by the step in the base air-fuel ratio during transients, and to reduce the P/I component, improving controllability. A learning control device for the base air-fuel ratio was devised to improve the performance of the catalyst and thereby reduce the cost of the catalyst.

That is, a map of learning correction coefficients α0 corresponding to engine operating conditions such as engine speed and load is provided in the RAM, and when calculating the injection amount Tt, the basic injection amount 'rp is corrected by α0 as shown in the following equation.

T i ='pp xCOEFXαxαo +Ts Then, the learning of α0 proceeds in the following steps.

i) In a steady state, detect the engine operating conditions and the control center value αC of α.

ii) Search for α0 that has been learned up to now corresponding to the engine operating conditions.

(2) Find the value α0+Δα/M from αC and α0, and update the memory with the result (learning value) as a new αO.

Note that Δα indicates the amount of deviation from the reference value α1, and Δα−α
C-α1, and the reference value α1 is generally 1.0.

Further, M is a constant.

By the way, in such a conventional learning method for air-fuel ratio feedback control, the deviation amount Δα cannot be detected accurately unless it is in a steady state, so learning is performed by detecting Δα only in a steady state. In this case, during a transient operating state, learning is not performed in an operating range where the vehicle operates only temporarily.

For this reason, as shown in Figure 1, for example, there are hunting areas where the learning progress is large (hereinafter referred to as learning areas) and other areas where learning progress is small (hereinafter referred to as unlearning areas). I end up. If the operating condition changes as shown by the arrow in this state, if a deviation occurs in the air-fuel ratio in the system, a deviation will occur in the correspondence between α and the air-fuel ratio λ between the learning region and the unlearning region. As a result, there is a difference in air-fuel ratio when transitioning between the learned area and the unlearned area,
This causes a problem in that exhaust emissions deteriorate in a transient state, and the learning effect is virtually ineffective.

<Purpose of the Invention> The present invention was made in view of such conventional problems, and it is possible to estimate the learned value in the unlearned area using the learned value in the learning area, and then use the estimated learning value to An object of the present invention is to provide a feed hack control device with a learning function that can improve control accuracy in transient operating conditions.

<Structure of the Invention> For this reason, as shown in FIG. 2, the present invention provides a control system that adjusts the control value of a controlled object operated under changing conditions according to the operating conditions in each different operating region so that the control value of the controlled object follows the target value. A control amount setting means for setting a control amount by adding or subtracting a feedback correction amount to a fixed control amount set by and a learning means for storing the learned value, further comprising a learning progress determining means for determining the degree of learning progress for each driving region by the learning means, and further comprising: A configuration including an auxiliary learning means for setting a learning value for a driving area of the learning progress variable vehicle based on the judgment from the judging means from a learning value in a learning progressing Keio University driving area around the said driving area by interpolation calculation. shall be.

<Example> Examples will be described below.

Figure 3 shows the bar 1 wear configuration.

1 is CPU, 2 is P-ROM, 3 is CMO for learning control
3-RAM, and 4 an address decoder. Furthermore, RAM
For No. 3, a backup power supply circuit is used to retain the memory contents even after the key switch is turned off.

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 the 02 sensor 8. There is an exhaust oxygen concentration signal from A and a battery voltage from battery 9, which are connected to analog input interface 10 and A
The signal is input via a /D converter 11.

12 is an A/D conversion timing controller.

Digital 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, for example, from the crank angle sensor 17 are inputted via a 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 from the CPUI (drive pulse signal to the fuel injection valve) is sent to the fuel injection valve 22 via the current waveform control circuit 21.
It is now sent to

Here, the CPU is running a program (RO
(stored in M2), performs input/output operations, arithmetic processing, etc., and controls the fuel injection amount.

Next, flowchart 1- in FIG. 4 will be explained.

Basic injection amount Tp (=
KxQ/N) is calculated.

In S2, various correction coefficients C0EF are set.

In S3, the update count C of the update count counter (increased in step 315, which will be described later) that counts the number of updates of the learning correction coefficient αO is compared with a predetermined value. After reducing the P/I portion of the λ control by a predetermined amount, the process proceeds to S5.

If it is less than the predetermined value, the process directly proceeds to S5 without changing P/r.

In S5, the output from the 02 sensor 8 is compared with the slice level, and the air-fuel ratio feedhack correction coefficient α is set by proportional-integral control based on the P/I component.

In S6, a voltage correction numerator s is set based on the battery voltage from the battery 9.

In S7, the current engine speed N and basic injection amount (load) Tp
The learning correction coefficient α0 in the corresponding driving range is searched from a map described later. A map of the learning correction coefficient α0 with respect to the engine speed N and the basic injection amount Tp is stored in the rewritable RAM 3, and all the maps are αo=l when learning has not started. Further, this map has approximately N=8 lattices and Tp=4 lattices.

88 to Sll are provided to detect a steady state, and in S8 a change in vehicle speed is determined based on the signal from the vehicle speed sensor 19, and in S9 the gear position is determined based on the signal from the neutral switch 15. , SIO determines the change in throttle opening based on the signal from the throttle sensor 6, and SII determines whether or not a predetermined time has elapsed. If it is within the predetermined time, the process returns to S8. In this way, if the change in vehicle speed is less than a predetermined value within a predetermined time, the gear is engaged, and the change in throttle opening is less than a predetermined value, it is determined that the steady state is present, and 312 to 314 The learning correction coefficient α0 is corrected. Additionally, 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 considered to be a transient state. Judgment, 31
The learning correction coefficient α0 is not modified in steps 2 to S14.

When the steady state is determined, the correction of the learning correction coefficient α0 in step 312 is performed based on the formula α0←αO+Δα/M, similar to the conventional one described above.

In S13, a new learning correction coefficient α0 is written into the corresponding operating region of the map in RAM3. That is, the data in RAMa is updated.

In S14, a count value C of an update number counter that counts the number of updates of the learning correction coefficient α0 in the current driving range is incremented.

In S15, the count value C of the update counter in the current operating region is compared with a predetermined value C+, and if the learning progress level, which is C2O4, is large, the process immediately proceeds to 818, where the injection amount Ti is calculated as described later. Ru.

If it is determined in step S15 that C<C+ and the learning progress is small, then in step 316 a search is made for a learning area where C≧C+ among the driving areas surrounding the unlearned area.

For example, in Fig. 1, when the driving area changes in the direction of the arrow, if it is in unlearned area a, the upper and lower learning areas A and B are searched on the map (if it is in unlearned area, learning areas A and B are searched). (areas A, B, D are searched).

Next, in S17, the learning correction coefficient αO in the learning areas searched as described above, for example, A and B, is read out.
Learning correction coefficient α0 in unlearned area a from these α0
is calculated by proportional interpolation.

Next, in 318, the injection amount Tt is calculated using the following equation.

T i = T p X COE F x ex x
An injection amount Ti is calculated at or above cx o + Ts, and a drive pulse signal corresponding to this injection amount Ti is given to the fuel injection valve 22 via the current waveform control circuit 21 at a predetermined timing.

As for the learning area, highly accurate injection amount control can be performed using the learning correction coefficient α0 that is actually learned during operation in the relevant area, as in the past, and in the unlearning area, it is based on the learning correction coefficients of the surrounding learning areas. Since injection amount control is performed using a highly reliable estimated learning correction coefficient determined by interpolation calculation, there is no difference in air-fuel ratio between the learning region and the unlearning region, and the exhaust emissions are reduced in transient conditions. Deterioration can be prevented and transient characteristics can be made smooth.

The estimated learning correction coefficient is stored in a map of RAM, and when the learning area becomes C≧C+ and changes to a learning area, the estimated learning correction coefficient stored in the map and the latest value of α are used. It may be updated as the learning correction coefficient α0.

<Effects of the Invention> As explained above, according to the present invention, an estimated learning value is determined by interpolation from the learning value of the learning progress area in the learning progress variable area, and the feet are calculated based on the estimated learning value. Because it is configured to perform hack control, stable control characteristics can be obtained even in transient operating conditions of the controlled object. For example, when applied to an air-fuel ratio feedbank control device for an internal combustion engine, it can eliminate steps in the air-fuel ratio and improve exhaust emissions. You can get great features such as

[Brief explanation of drawings]

Fig. 1 is a graph for distinguishing between a learned area and an unlearned area, Fig. 2 is a block diagram showing the configuration of the present invention, Fig. fA3 is a hardware configuration diagram showing an embodiment of the present invention, and Fig. 4 is the same. ”
This is a flowchart showing the control process on the two side. 1...CPtJ 3...CMO3-RA for learning control
M5...Air flow meter 8...02 sensor 1
7...Crank angle sensor 22...Fuel injection valve Patent applicant Japan Electronics Co., Ltd. Agent Patent attorney Fujio Sasashima (I5) Figure 1 Figure 2

Claims (1)

[Claims] The control amount is set by adding or subtracting the feedback correction amount to the fixed control amount that is set according to the operating conditions for each different operating region so that the control value of the controlled object that is operated under fluctuating conditions follows the target value. and a learning means for performing learning to correct and update the fixed control amount in order to reduce the deviation amount between the control center value of the feedbank correction amount and the reference value, and storing the learned value. In the feedback control device with a learning function, there is provided a learning progress determining means for determining the degree of learning progress for each driving region by the learning means, and a learning progress determining means for determining the learning progress for each driving region of the learning progress variable vehicle based on the determination from the determining means. A feedback control device with a learning function, characterized in that it is provided with auxiliary learning means for setting a learning value by interpolation calculation from a learning value in a driving area around the driving area where the learning progresses at Keio University.