JP2803364B2 - Digital controller - Google Patents

Description

Description: TECHNICAL FIELD The present invention relates to a digital control device, and in particular, considers a system for controlling an idle speed of an internal combustion engine as a dynamic system in consideration of an internal state thereof, and The present invention relates to a device that controls an idle speed in accordance with a control value obtained according to a state variable that defines an internal state.

2. Description of the Related Art Conventionally, as an idling speed control device of this kind for an internal combustion engine, it is disclosed in, for example, JP-A-59-145339. In this publication, feedback control is performed from an engine state in which the idling speed is not feedback-controlled. This figure shows a configuration devised so that the engine speed smoothly reaches the target speed when the engine state is shifted.

However, in the configuration disclosed in the above publication, when it is determined that the feedback control of the rotation speed is to be started, the state variable amount and the detected rotation speed and the target rotation speed used for obtaining the control value for the control of the rotation speed are determined. Since each initial value of the integral term of the deviation is given according to the detected rotation speed when the throttle valve is fully closed and when it is determined that feedback control is started, the throttle valve is fully closed. The initial values of the state variable amount and the integral term must be provided corresponding to the detected rotational speeds when the feedback control is started and when the feedback control is determined to be started. Even if it is finely set, the internal state of the system that controls the idle speed of the internal combustion engine is not determined uniquely by the engine speed, so the given initial value A completely or truly suitable are not known, in some cases there is a problem that there is a possibility that exacerbate convergence to the target rotational speed.

Accordingly, the present invention has been made in view of the above-described problems, and a first object is to reduce the state in which the actuator is driven at a value irrelevant to the internal state of the internal combustion engine. It is an object of the present invention to provide a digital control device that improves convergence to a target rotation speed when a transition is made to a state driven by a value set according to an internal state of an internal combustion engine.

Conventionally, when a load disturbance of an air conditioner or the like enters, a value calculated by feedback control is adjusted by a correction amount corresponding to the disturbance to improve a transient response of an idle speed.

By the way, in the idle speed control device during the internal combustion period,
The control value (duty ratio of the duty signal for controlling the control solenoid) that is finally output is determined in consideration of the movable range and the like of an actuator (for example, an idle speed control valve) for controlling the idle speed. (0% to 100%) is generally performed. Further, as shown in JP-A-59-43942, a control value obtained at a previous calculation timing used for determining a control value finally output,
In order to prevent the state variable amount determined from the detected rotation speed or the integral value of the deviation between the target rotation speed and the detected rotation speed from becoming excessively large or small, and adversely affecting the operation of the actuator, Upper and lower limits are also set for the integral value and the state variable amount.

For this purpose, the configuration described in the prior art, in which the value calculated by the feedback control with the correction amount according to the disturbance is applied to the above-described device having upper and lower limits for the final control value, the integral value, and the state variable amount. In this case, even if the detected rotation speed is higher than the target rotation speed and the final control value is minimized, a correction amount is added if a load disturbance is present at that time, and the predetermined amount is smaller than the lower limit of the final control value. May not be less than the large value. In such a case, since the idle speed control valve is kept open by a predetermined amount, the detected rotation speed may not be converged to the target rotation speed and may be maintained at a high rotation speed. .

Therefore, the second object of the present invention is to correct the control value in a feed-forward manner in consideration of the influence of load disturbance and the like, and to reduce the detected rotation speed to the target rotation speed very accurately without the above-described problem. An object of the present invention is to provide a digital control device improved so as to be controllable.

Further, a third object of the present invention is to provide a digital control device capable of constructing a high-precision and low-order dynamic model and realizing extremely accurate control based on the constructed dynamic model. It is in.

DISCLOSURE OF THE INVENTION Accordingly, in the present invention, the state variable amount is always set even while the current control value is set independently of the current state variable amount.

As a result, even when the current control value shifts from a state where it is set independently of the current state variable amount to a state where it is set according to the current state variable amount, the current state variable amount is already set. Therefore, the convergence to the target control value is greatly improved.

In the present invention, when a change in the state of the load is detected,
At that time, the integrated value is corrected in a feedford manner before being limited to a predetermined range according to the state change, and the integrated value can fully fluctuate within the predetermined range, and the feedforward correction according to the load disturbance is performed. Even so, the control value can be made to fully change within a predetermined range. Therefore, even if the detected rotational speed is higher than the target rotational speed and the final control value is minimized as in the prior art described above, a correction amount is added if a load disturbance is present at that time, and the final control value And the idle speed control valve is kept open by the predetermined amount, and the detected rotation speed cannot be converged to the target rotation speed. Can be solved, and the detected rotation speed can be accurately converged to the target rotation speed.

In configuring the present invention, the dynamic model for the control system is divided into a dead time portion and a portion after the dead time in consideration of the dead time, and the dynamic model is identified by a discrete system for each. ing. Then, a dynamic model of the entire control system is constructed in advance based on these identified dynamic models. By constructing the dynamic model in consideration of the dead time, a highly accurate dynamic model can be obtained. In the present invention, the state variable amount is determined according to the control input / output amount based on the pre-established dynamic model, and the control input amount is output based on the state variable amount. (Using a low-order dynamic model), it is possible to realize a highly accurate control.

Further, by applying the present invention to idle speed control of an internal combustion engine, high-precision idle speed control can be realized using a low-order dynamic model, that is, with a simple configuration.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram of the present invention, FIG. 2 is a schematic diagram showing an internal combustion engine to which an embodiment of the present invention is applied, and peripheral devices thereof, and FIG. FIG. 4, FIG. 4 is a characteristic diagram of the rotation speed with respect to the control signal, FIG. 5, FIG. 6 is a block diagram of a system for controlling the idle rotation speed, and FIG. 11 to 13 are flowcharts of an ISC valve control program in the case of two integrators,
FIG. 14 to FIG. 28 are flowcharts showing the contents of the main part of another embodiment.

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments will be described below. The blocks in this embodiment are:
As shown in FIG. 1, a rotational speed detecting means for detecting the rotational speed of the internal combustion engine, a rotational speed adjusting means for adjusting the rotational speed of the internal combustion engine, and the detected rotational speed detected by the rotational speed detecting means are desired. Control means for calculating a control value for controlling the rotation speed adjusting means so as to coincide with the target rotation speed at predetermined intervals, and outputting a control signal according to the control value. In the rotation speed control device, the control means, the detected rotation speed and the state variable amount set at the previous calculation timing and the control value corresponding to the control signal output to the rotation speed adjustment means, A first control value setting means for setting a state variable amount according to the state variable amount, and setting a current control value according to the state variable amount; and a state variable amount set at the previous calculation timing and the previous calculation timing. For the rotation speed adjusting means A second control value setting means for setting a state variable amount according to the control value corresponding to the control signal output as described above, and setting the current control value to a value irrelevant to the state variable amount; There is provided a selecting means for selecting one of the first control value setting means and the second control value setting means according to a state of the internal combustion engine.

Hereinafter, an idle speed control device to which the present invention is applied will be described with reference to the drawings. FIG. 2 is a schematic configuration diagram showing the engine 10 and its peripheral devices for performing idle speed control described below. As shown in the figure, in this embodiment, each control of the ignition timing, the fuel injection amount, and the idle speed of the engine 10 is performed by the electronic control unit 20.
Here, the control of the idle speed will be mainly described.

The engine 10 is mounted on a vehicle, and is a four-cylinder four-cycle spark point type as shown in FIG. 2, and its intake air is supplied from an upstream air cleaner 21, an air flow meter 22, an intake pipe 23, The fuel is sucked into each cylinder via a surge tank 24 and an intake branch pipe 25, while the fuel is pressure-fed from a fuel tank (not shown) and a fuel injection valve 26 provided in the intake branch pipe 25.
a, 26b, 26c, 26d. In addition, the engine 10 transmits the high-voltage electric signal supplied from the ignition circuit 27 to the ignition plugs 28a, 28b, 28c, 28 of each cylinder.
(d) a distributor 29 for distributing the engine d, a rotational speed sensor 30 for detecting the rotational speed Ne of the engine 10 provided in the distributor 29, a throttle sensor 32 for detecting the opening degree TH of the throttle valve 31, and a cooling water temperature Thw for the engine 10. And an intake air temperature sensor 34 for detecting the intake air temperature Tam. The rotation speed sensor 30 is provided to face a ring carrier that rotates in synchronization with the crankshaft of the engine 10, and the engine speed is proportional to the rotation speed.
A pulse signal of 24 pulses is output at one rotation of 10, that is, at 720 ° C. The throttle sensor 32 is used to open the throttle valve TH.
And an on / off signal from an idle switch for detecting that the throttle valve 31 is almost fully closed.

On the other hand, the intake system of the engine 10 has a throttle valve 31
Of the intake air when the engine 10 is idling
A bypass passage 40 for controlling AR is provided. The bypass passage 40 includes air conduits 42 and 43 and an air control valve (hereinafter, referred to as an ISC valve) 44. This ISC valve 4
Reference numeral 4 denotes a proportional solenoid (linear solenoid) control valve which varies the area of the air passage between the air conduits 42 and 43 depending on the position of the plunger 46 movably set in the housing 45. To control. ISC valve 44
Normally, the plunger 46 is set in a state where the air passage area is reduced to zero by the compression coil spring 47.
46 is configured to be driven to open the air passage.
That is, the flow rate of the bypass air is controlled by continuously changing the exciting current to the exciting coil 48. In this case, the exciting current for the exciting coil 48 is
It is controlled by performing so-called pulse width modulation PWM for controlling the duty ratio of the pulse width applied to the excitation coil 48.

The ISC valve 44 is driven and controlled by the electronic control unit 20 in the same manner as the fuel injection valves 26a to 26d and the ignition circuit 27.
A valve or the like by step motor control is appropriately used.

The electronic control unit 20 includes a well-known central processing unit (CPU) 52 and a read-only memory (RO)
M) 52, a random access memory (RAM) 53, a backup RAM 54, etc., are configured as an arithmetic and logic operation circuit, and an input port 56 for inputting from each sensor described above.
And output port 58 that outputs control signals to each actuator
And the like, and are interconnected via a bus 59. The electronic control unit 20 receives, via an input port 56, an intake air amount AR, an intake temperature Tam, a throttle opening TH, a cooling water temperature Thw, and a rotation speed Ne.
, And the fuel injection amount τ, ignition timing
Iq, the ISC valve opening θ, and the like are calculated, and control signals are output to each of the fuel control valves 26a to 26d, the ignition circuit 27, and the ISC valve 44 via the output port 58. Among these controls, the idle speed control will be described below.

The electronic control unit 20 is designed in advance by the following method in order to perform idle speed control.

(1) Modeling (Identification) of Controlled Object In this case, an auto-regressive movement of order [n, m] having a dead time P (= 0,1,2. The approximation is performed using an average model and further considering a disturbance d. First, the model of a system for controlling the number of idle rotations using autoregressive moving average _{model, Ne (i) = a 1} · Ne (i-1) + a 2 · Ne (i-2) + ...... + a n · can be approximated by Ni (i-n) + b 1 · u (i-1-p) + b 2 · u (i-2-p) ...... + b m · u (i-m-p) ...... (1), further merits model of the controlled system in view of the disturbance _{d Ne (i) = a 1} · Ne (i-1) + a 2 · Ne (i-2) + ...... + a n · Ni (i-n) + b ₁ · u (i−1−p) + b ₂ · u (i−2−p) ··· + b _m · u (im−p) + d (i−1) (2)

In the present embodiment, a system for controlling the idling speed is used as an autoregressive moving average model of order [2,2] with n = m = 2, and the delay p due to the sampling time (dead time) is p = As 2, Ne (i) = a ₁ · Ne (i−1) + a ₂ · Ne (i−2) + b ₁ · u (i−3) + b ₂ · u (i−4) (3) obtain. Furthermore a system model for controlling the idle speed by considering the disturbance d _{to, Ne (i) = a 1} · Ne (i-1) + a 2 · Ne (i-2) + b 1 · u (i- 3) + b ₂ · u (i−4) + d (i
-1) (4) Here, u represents the control value of the ISC valve 44, and in this embodiment, corresponds to the duty ratio of the pulse signal applied to the exciting coil 48. Also, i
Is a variable indicating the number of controls from the start of the first sampling.

For the model approximated in this way, the transfer function G of the system for controlling the idle speed using the step response is obtained, and from this, it is possible to experimentally determine the constants a ₁ , a ₂ , b ₁ , b ₂ of the above model. Easy. By determining the constants a ₁ , a ₂ , b ₁ , and b ₂ , the model of the system that controls the idle speed is determined.

Next, a procedure for identifying a further improved control object will be described with reference to a flowchart shown in FIG.

First, in step 101, the step response of the control target is observed. A duty ratio signal D ₁ = D ₀ + ΔD (D ₀ is a current duty ratio, and ΔD is a duty ratio for increasing the ISC valve opening by Δθ) to increase the ISC valve opening by a predetermined opening Δθ. Then, the behavior of the rotation speed at that time is measured. Behavior of the rotational speed, as shown in FIG. 4, when a signal of the duty ratio D ₁ is increased by a predetermined opening degree Δθ of the ISC valve opening is output, the rotation speed Ne is delayed by the dead time L increases Start to do.

Next, in step 102, the model is separated. According to the dead time L measured in step 101, the dynamic model of the control system is separated into a dead time L part and a part after the dead time L.

Then, in step 103, a partial identification of the dead time L as the first dynamic model is performed. Here, the transfer function G _(S) when the dead time L is model-identified in a continuous system is And infinite order. However, the sampling period Δ
By setting t to Δt = L / N (where is an arbitrary integer), the transfer function G _a1 (z) when the dead time L is identified by a discrete system is G _a1 (z) = 1 / z ^N Become. In this embodiment, L = 240 _msec , N = 2, and Δt = _240/2 = 120 _msec .

That is, G _a1 (z) = 1 / z ₂ ... Therefore, the first dynamic model becomes a simple transfer function as in the following equation in a discrete system.

Next, in step 104, model identification of a portion after the dead time L as the second dynamic model is performed. A dynamic model of a discrete system in which the sampling period is Δt described above can be represented by a transfer function G _a2 (z) = (b ₁ z + b ₂ ) / (z ² + a ₁ z + a ₂ ).

Here, the constants a ₁ , a ₂ , b ₁ , and b ₂ of the transfer function G _a2 (z) can be experimentally obtained by the least square method or the like.

Finally, at step 105, step 103 and step
Based on the transfer functions G _a1 (z) and G _a2 (z) for the dynamic model respectively identified in 104, the transfer function G _a (z) of the dynamic model of the entire system is constructed. Becomes

Here, G _a (z) = Ne (z) / u (z). Here, Ne (z) is a function of the rotational speed which is the control output amount, and u
(Z) is a function of the duty ratio of the pulse signal, which is the control input amount. Therefore, the following equation can be rearranged with respect to Ne (z): Ne (z) = (a ₁ z ⁻¹ + a ₂ z ⁻² ) × Ne (z) + (b ₁ z ⁻³ + b ₂ z ⁻⁴ ) × u (z )… '. Here, z ^-1 represents a time delay operator.

Therefore, the estimated rotation speed Ne (i) Ne (i) than the 'equation _{= a 1 · Ne (i-} 1) + a 2 · Ne (i-2) + b 1 · u (i-3) + b 2 · u (I-4) ... can be estimated by the following equation. When the external d is considered in the equation, the estimated rotational speed Ne (i) is expressed as Ne (i) = a ₁ · Ne (i−1) + a ₂ · Ne (i−2) + b ₁ · u (i−3) + B ₂ · u (i−4) + d (i−1)...

Here, u indicates the control input amount of the ISC valve 44, and in this embodiment, corresponds to the duty ratio of the pulse signal applied to the exemplary coil 48. Further, i is a variable indicating the number of times of control since the start of the first sampling.

In FIG. 3 step 104, the seeking dynamic model of direct discrete system, the dynamic model of continuous system becomes ^{_{G (s) = Kω m 2}} / (S 2 + 2Tω m S + ω m 2). Here, assuming that the rotation increase amount ΔNe ₁ , the overshoot amount ΔNe ₂ , the overshoot occurrence time t _ov , and the ISC valve change amount Δθ It is.

The transfer function G _(s) of the continuous system may be discretized at the sampling period Δt to obtain the transfer function Gaz (z) of the discrete system.

Although the controlled object is described as a system that controls the idle speed, this modeling method is a system in which the control output amount changes after a certain time (dead time) after the control input amount is input to the actuator. Then, it is applicable to any control system.

As described above, in the improved modeling method, the dynamic model for the control system is divided into a dead time portion and a portion after the dead time in consideration of the dead time, and the dynamic model is discretely processed for each. Have been identified. Then, a dynamic model of the entire control system is constructed in advance based on these identified dynamic models. By constructing the dynamic model in consideration of the dead time, a highly accurate dynamic model can be obtained. Since the state variable amount is determined based on the control input / output amount based on the pre-established dynamic model and the control input amount is output based on the state variable amount, a simple configuration (low order (Using the dynamic model described above), there is an excellent effect that highly accurate control can be realized.

Further, by applying this method to idle speed control of an internal combustion engine, high-precision idle speed control can be realized using a low-order dynamic model, that is, with a simple configuration.

Furthermore, in the identification of the dynamic model, the sampling period is set to 1 / N of the dead time (N is an arbitrary integer), and the dynamic model is identified in a discrete system. Model can be identified. Therefore, the operation formula used for the control operation in the control device is simplified, and the effect that the operation load can be reduced is produced.

Equation (4) Rewriting with Get. Therefore, _{Is, X 1 (i) = Ne} (i), X2 (i) = Ne (i-1), X 3 (i) = u (i-1), X 4 (i) = u (i-2) , X ₅ (i) = u (i−3) (6)

(3) Design of regulator A general optimal regulator does not have a function of converging an output to a target value. Therefore, in the case of the idle control, the error (e (i) = NT) between the target rotation speed and the actual rotation speed is obtained.
(I)-An expansion system regulator that introduces Nei) is required. Therefore, in this embodiment, Aim to be.

Also, as shown in JP-A-64-8336,
If only one integrator is provided, when the target rotational right changes stepwise (for example, air conditioner ON / OFF switching, shift change between ND, etc.), it converges without generating a deviation from the target rotational speed. In addition, when the target rotation speed changes on a ramp (for example, at the time of warm-up, at the time of starting, or after racing), there is a problem that the control is performed with a certain amount of deviation. Therefore,
In the present embodiment, it is intended to execute the calculation by two integrators when the target rotation speed changes by a ramp, and to execute the calculation by one integrator when the target speed does not change or does not change.

(I) When the disturbance d (i) and the target rotational speed NF (i) are constant From the conditions, (1-q ^-1 ) d (i) = 0, (1-q ^-1 ) NF (i) = 0 ( (q is a time transition operator) Therefore, when (1-q- ¹ ) is applied to e (i + 1) = NF (i + 1) -Ne (i + 1), (1-q- ¹ ) e (i + 1) = (1- q ⁻¹ ) (NF (i + 1) −Ne (i + 1)) = − (1−q ⁻¹ ) Ne (i + 1) where (1−q ⁻¹ ) Ne (i + 1) = a ₁ (1−q ⁻⁾ in _{^{1) x 1 (i) +}} a 2 (1-q -1) x 2 (i) + b 1 (1-q -1) x 4 (i) + b 2 (1-q -1) x 5 (i) Therefore, the following expansion system is obtained.

When the optimal regulator is used in the equation, (1−q ⁻¹ ) u (i) = K ₁ (1−q ⁻¹ ) × ₁ (i) + K ₂ (1−q ⁻¹ ) × ₂ (i) + K ₃ (1-q ^-1) x ₃ becomes _{(i) + K 4 (1} -q -1) x 4 (i) + K 5 (1-q -1) x 5 (i) + K a e (i).

Dividing both sides by (1-q ^-1 ), Where 1−q ⁻¹ Then, u ₁ (i) = u ₁ (i−1) + Ka (NF−Ne (i)) (9)

Therefore, ISC control value u of the valve 44 (i) _{is, u (i) = K 1} Ne (i) + K 2 Ne (i-1) + K 3 u (i-1) + K 4 u (i-2) + K ₅ u (i−3) u ₁ (i) (8) In addition, Is an optimal feedback gain, and Ka is an integration constant.

(Ii) When the disturbance d (i) and the target rotation speed NF (i) change in a ramp shape When the target rotation speed changes in a ramp shape, NF (i) −NF (i−1) = NF (i) −1) −NF (i−2) (constant slope) That is, (1−q ⁻¹ ) d (i) = 0, (1−q ⁻¹ ) ² NF
(I) = 0 Therefore, when (1-q ⁻¹ ) ^{2 is applied} to e (i + 1) = NF (i + 1) −Ne (i + 1), (1−q ⁻¹ ) ² e (i +
1) = (1−q ⁻¹ ) ² (NF (i + 1) −Ne (i + 1)) = − (1−q ⁻¹ ) ² Ne (i + 1) where (1−q ⁻¹ ) ² Ne (i + 1) ) = A ₁ (1-q ^-1 ) ² x ₁ (i) + a ₂ (1-q ^-1 ) ² x ₂ (i) + b ₁ (1-q ^-1 ) ² x ₄ (i) + b ₂ ( 1−q ⁻¹ ) ² x ₅ (i), the following expanded system is obtained.

∵ (1−q ⁻¹ ) ² e (i + 1) = − a ₁ (1−q ⁻¹ ) ² x ₁ (i) −a ₂ (1−q ⁻¹ ) ² x ₂ (i) −b ₁ ( 1-q ^-1 ) ² x ₄ (i) -b ₂ (1-q ^-1 ) ² x ₅ (i) (1-2 q ^-1 + q ^-2 ) e (i + 1) = -a ₁ (1-q ^-1 ) ² x ¹ (i) -a ₂ (1-q ^-1 ) ² x ₂ (i) -b ₁ (1-q ^-1 ) ² x ₄ (i) -b ₂ (1-q ^-1) ) ² x ₅ (i) e (i + 1) = -a ₁ (1-q ^-1 ) ² x ₁ (i) -a ₂ (1-q ^-1 ) ² x ₂ (i) -b ₁ (1- _{^{q -1) 2 x 4 (i}} ) -b 2 (1-q -1) 2 x 5 (i) + 2e (i) -e (i-1) the use of optimal regulator in formula, (1-q ^{- 1) 2 u (i) =} K 1 (1-q -1) 2 x 1 (i) + K 2 (1-q -1) 2 x 2 (i) + K 3 (1-q -1) 2 x 3 (i) + K ₄ becomes ^{(1-q -1) 2 x} 4 (i) + K 5 (1-q -1) 2 x 5 (i) + K a e (i) + K b e (i-1). Dividing both sides by (1-q ^-1 ) ² gives Becomes

here, ^{When, (1-2q -1 + q -2)} u 1 (i) = e (i) _{i.e., u 1 (i) -2u 1} (i-1) + u (i-2) =
e (i).

Therefore, u ₁ (i) = 2u ₁ (i−1) −u ₁ (i−2) + e (i) (9) ′ Therefore, the control value u (i) is expressed as u (i) = K _{1 Ne (i) + K 2 Ne} (i-1) + K 3 u (i-1) + K 4 u (i-2) + K 5 u (i-3) + K a u 1 (i) + K b u 1 (i- 1) It is obtained as (8) '.

5 and 6 are block diagrams of a system for controlling the idle speed modeled as described above. In FIGS. 5 and 6, the control value u (i-1) is used.
Is derived using the Z ^-1 transform to derive from u (i),
This corresponds to storing the past control value u (i-1) in the RAM 53 and reading and using it at the next control.

In the state where the block P1 surrounded by the dashed line in FIGS. 5 and 6 is performing feedback control of the rotation speed to the target rotation speed, , The block P2 determines the integral term u ₁ (i) (accumulation unit), and the block P3 is determined by the block P1. And a part for calculating the current control value u (i) from the integral term u ₁ (i) obtained in the block P2.

And the determination of the integration constants Ka and Kb And the integration constant KaKb can be determined, for example, by the following method.

(Optimal servo system) And the integration constants Ka and Kb are evaluation functions, Is minimized. Here, the evaluation function J is intended to minimize the deviation of the idle speed Ne (i) as the control output from the target speed NF while restricting the movement of the control value u (i) of the ISC valve 44. This is intended, and the weighting of the constraint on the control value u (i) can be changed by the values of the weighting parameters Q and R. Therefore, the simulation is repeated until the optimal control characteristics are obtained by variously changing the values of the weight parameters Q and R, Should be determined.

And the above Depends on model constants a ₁ , a ₂ , b ₁ , and b ₂ . Therefore, in order to guarantee the stability (robustness) of the system with respect to the fluctuation (parameter fluctuation) of the system that controls the actual idle speed, the model constants a ₁ and a
₂ , b ₁ , b ₂ It is necessary to design the integral coefficients Ka and Kb. Therefore, the simulation satisfies the stability by taking into account the possible variations of the model constants a ₁ , a ₂ , b ₁ , b ₂ And the integration constants Ka and Kb. Factors of change are ISCV
In addition to a change with time such as settling of the valve 44 and clogging of the bypass passage, a change due to a load change or the like can be considered. Note that this And, Ka, Kb may be provided in advance, for example, two types, one corresponding to a small load fluctuation state and one corresponding to a large load fluctuation state, and may be switched according to the load fluctuation state. Is also conceivable.

The modeling of the controlled object, the method of displaying the state variable amount, the design of the regulator, and the determination of the optimal feedback gain have been described above. However, these are determined and obtained in advance, and the result, that is, Actual control is performed using only the equations (8) and (9) or the equations (8) 'and (9)'.

In the present embodiment, the feedback processing using the equations (8) and (9) or the equations (8) 'and (9)' is performed only when the state of the engine 10 satisfies a predetermined feedback execution condition. If the feedback execution condition is not satisfied (open state), the (8),
The processing using the equation (9) or the equations (8) 'and (9)' is not executed inside the electronic control unit 2, and the control value for the ISC valve 44 is determined according to other predetermined processing. Further, in this embodiment, in the open state, the processing for the next feedback processing is executed at each calculation timing for determining the control value.

The processing contents executed by the CPU 51 of the electronic control device 20 in consideration of the open processing in the open state in addition to the above-described feedback processing will be described with reference to FIGS. 7, 8, 9, and 10. I do.

7 is a control program of the ISC valve 44, which is executed by interruption every predetermined time (for example, every 100 msec) while an IG switch (not shown) is closed.

First, when the process is started by an interrupt, it is determined in step 302 whether 3 seconds have elapsed after the start of the engine 10 is completed. This is for controlling from a state where it is recognized that the engine has escaped from an unstable state immediately after the engine is started. The start of the engine 10 is completed, for example,
When the rotation speed Ne exceeds 500 rpm, it is determined that the start is completed.

If it is determined in step 302 that 3 seconds have elapsed after the start is completed, the routine proceeds to step 304, where it is determined whether the throttle valve 31 is fully closed and the idle switch is on (LL: ON). If it is determined in step 304 that LL is ON, the process proceeds to step 306, where it is determined whether or not warm-up is completed.
If the warm-up is completed, the process proceeds to step 308.

In step 308, the flag (F / B flag) that is set to 1 while the feedback (F / B) process is being executed is 1
Is determined, and if the F / B flag = 1, the routine proceeds to step 310. In step 310, it is determined whether or not the target value lifting amount NFOPEN set immediately after shifting from the open state to the state in which the feedback processing is executed is less than 5 rpm.
If NFOPEN <5 rpm, the lift amount NFOPEN is set to 0 in step 312, and the process proceeds to step 314. Also, if NFOPEN ≧ 5 rpm, it is determined whether 1 second has elapsed since the transition to the F / B state in step 316 and the F / B processing has been started, and if not elapsed, the process proceeds to step 314. 5 rp lift NFOPEN
After correcting the value to a value smaller by m (NFOPEN ← NFOPEN-5 rpm), the process proceeds to step 314. In step 314, the reference rotation speed NFB
The target rotation speed NF is determined by adding the lifting amount NFOPEN to (for example, 700 rpm).

In step 320, an F / B process described later is executed corresponding to the target rotational speed NF determined in step 314.

On the other hand, if it is determined in step 308 that the F / B flag is 0, the process proceeds to step 322, where the latest rotation speed Nen and the reference rotation speed NFB obtained based on the signal of the rotation speed sensor 30 are set to a predetermined value NA. (For example, 200 rpm), Nen ≦ N
If FB + NA, the process proceeds to step 324. If Nen> NFB + NA, the process proceeds to step 326. In step 326, LL: 3 seconds after ON
It is determined whether it has elapsed, and if it has elapsed, the process proceeds to step 324.

In step 324, the F / B flag is set to 1 and the routine proceeds to step 328. In step 328, the lifting amount NFOPEN is obtained by subtracting the reference rotational speed NFB from the latest rotational speed Nen. Therefore, F / B
As the initial value of the target rotation speed NF at the start of the process, the rotation speed at the time of determining to start the F / B process is set.

Also, in step 302, if 3 seconds have not elapsed after the start, or if LL: OFF in step 304, or if the warm-up is not completed in step 306, or if 3 seconds have not elapsed since LL: ON in step 326, Go to step 330. At step 330, the F / B flag is set to 0, and at step 332, an open process described later is executed.

After completing the processing in step 320 or step 332,
In step 334, a storage process described later for the next feedback process is executed, the control program is temporarily ended, and the process proceeds to another engine control program.

The flowchart in FIG. 8 shows a case where the disturbance and the target rotation speed are constant in the F / B processing in step 320, and the processing based on the above equations (8) and (9) is executed. More specifically, in step 402, the latest rotational speed Nen is substituted for Ne (i), and in step 404, the calculation of the above-mentioned equation (9) is executed to obtain u ₁
(I) is obtained, and then, in step 406, the calculation of Expression (8) is executed to obtain the current control value u (i). Then, a control signal having a duty ratio corresponding to the current control value u (i) thus obtained is output from the output port 58 to the ISC valve 44.

That is, the latest rotational speed Nen is set to Ne (i) for calculation, and the deviation between Ne (i) and the target rotational speed NF is obtained by the previous process and is stored in the RAM 53 as the integral term u. ₁ (i
In addition to -1), u ₁ (i) of this time is determined. Then, the set Ne (i) and the previous state variable Ne (i−i) stored in the RAM 53 in preparation for the current F / B processing in the previous processing.
1), u (i-1), u (i-2), u (i-3), etc., and the current state variable [Ne (i) Ne (i-1) u (i-
1) u (i-2) u (i-3)] is determined, a matrix operation is performed on the current state variable amount and the optimal feedback gain, and Ka · u ₁ (i) is further added to control the current control value u. (I)
Has been established.

FIG. 9 shows a flowchart of the open process in step 332. In this open process sets the current control value u (i) to a predetermined value u ₀ in step 502. The predetermined value u ₀ may be an arbitrary constant value such as 100%, 0% or 50% as a duty ratio, or may be a value determined according to a detection parameter such as a cooling water temperature Thw.

In step 504, the latest rotational speed Nen is substituted for Ne (i). And in step 506, it was set in step 504
Ne (i) and Ne (i-1), u (i) stored in the RAM 53
-1), u (i-2), u (i-3) and the current control value u (i) set in step 502, based on the expression (8), the control value u (i) set this time. And the integral term u ₁ (i) that matches the current state variable quantity is inversely calculated.

Note that the state variable amount at the time of this open processing is Ne (i) set in step 540 and Ne (i-1), u (i-1), u (i-2), u stored in the RAM 53. (I-3)
From [Ne (i) Ne (i-1) u (i-2) u (i
-2) u (i-3)].

In step 508, a control signal of the duty ratio is output from the output port 58 to the ISC valve 44 according to the current control value u (i) set in step 502.

Next, the storage processing in step S334 will be described with reference to the flowchart in FIG. In this process, first, in step 602, Ne (i), u (i-2) in the state variable amount set in one of step 320 (F / B processing) and step 332 (open processing) executed immediately before. , u (i-1) to Ne (i-1), u (i-3), u (i-2), respectively, and the current control value u ( i) and u ₁ (i) are each u
(I-1), is substituted for u _{1 (i-1).}

Next, in step 604, Ne (i
-1), u (i-3 ), u (i-2), u (i-1), u 1 (i
-1) is stored in the RAM 53.

That is, in the storage process, the Ne used in steps 320 and 332 is used.
(I), u (i- 2), u 1 in the next F / B processing and the next open process using u (i-1) and the control value u which defines the same step (i) of (i) The stored state variable amount is updated and stored in preparation for the inverse operation. Also, u ₁ (i) determined in step 320 (F / B processing) is stored in preparation for the next F / B processing. Further, u ₁ (i) calculated in step 332 (open processing) is also stored as an initial value at the time of calculating u ₁ (i) by equation (9) in the next F / B processing. Moreover, in this embodiment, the data is stored after being changed to the form used in the processing at the next calculation timing (step 602).

Therefore, according to the control program of the ISC valve 44,
While in the open state, the state variable amount stored in the RAM 53 is updated based on the number of revolutions in the open state and the control value for the ISC valve 44. since the initial value of u ₁ from the state variable quantity and the control value is calculated in preparation for the next F / B processing, the state used in the F / B processing when moved from the open state to the F / B conditions the initial value of the variable quantity and u ₁ has a a representation of a state of the system that controls the number of idle rotation in the open state immediately before. Therefore, the change in the rotation speed immediately after the shift to the F / B state can be made extremely smooth. In addition, since the optimal state variable amount is determined immediately after the start of the F / B, it is possible to quickly converge to the target rotational speed.

In the above-described processing, the initial value of the nominal rotation speed NF when the F / B processing is started by shifting from the open state to the F / B state is set to F / B.
The processing is performed so as to be the actual rotation speed Nen at the time when it is determined that the processing is started.

For example, at the start of F / B processing, if the target rotation speed is determined regardless of the actual rotation speed at that time, a deviation between the target rotation speed and the actual rotation speed will occur from the start of F / B processing, and in some cases, In some cases, the actual rotation speed is larger by 400 to 500 rpm, and when the F / B processing is started in such a state, the ISC valve 44 is suddenly closed to lower the actual rotation speed according to a large deviation. Controlled. When an engine load is applied, such as when the on-vehicle air conditioner is turned on while the actual rotation speed is being reduced to the target rotation speed, the speed at which the rotation speed decreases is promoted, and the actual rotation speed is reduced to the target rotation speed. There is a risk of undershooting larger than the number and causing engine stall in some cases, but such problems can be solved by setting the initial value of the target rotation speed at the start of F / B as described above Become like In the above process, the lift NFOPEN at the start of F / B is calculated by comparing the actual rotation speed Nen and the reference rotation speed NF.
Since the lifting amount NFOPEN is held for 1 second after the start of F / B, for example, when the vehicle shifts from deceleration operation to the idle state and enters the F / B state In the state where the number of revolutions is reduced to the idling revolution, the rate of decrease of the number of revolutions is suppressed, and a sharp drop in the number of revolutions can be prevented.

Furthermore, since the lifting amount NFOPEN is decreased by a predetermined value until it becomes zero one second after the start of F / B, the target rotation speed
The actual rotation speed smoothly follows the decrease in NF.
It will begin to decrease.

That is, according to the above-described processing, the number of revolutions when the state shifts from the open state to the F / B state shows an extremely smooth behavior, an extremely stable idle state is obtained, and the drivability is markedly improved. Can be improved.

In the above, FIGS. 8, 9, and 10 have described the case where the disturbance and the target rotation speed are constant or change stepwise (Equations (8) and (9)), but the disturbance and the target rotation speed change in a ramp shape. For the case (Equations (8) ', (9)'), the eleventh,
This is shown in the flowchart of FIG.

As described above, when the target rotation speed does not replace the previous time or changes stepwise, the calculation is performed by one integrator, and when the target rotation speed changes in a ramp shape, the calculation is performed by two integrators. Therefore, a deviation from the target rotation speed can be eliminated.

By the way, in the storage processing of step 334 in the processing of FIG. 7, the state variable amount and the integral term are stored in a form corresponding to the next processing, but the state variable amount is stored in the form used in the arithmetic processing in steps 320 and 332. , Integral terms may be stored and converted into state variable quantities corresponding to the computation at that time in the processing of steps 320 and 332 of the next processing.

Specifically, as shown in FIGS. 14, 15, and 16, the F / B
In the processing, the previous state variable amounts Ne (i), u (i-2), u (i-1), u (i) stored in step 702 are respectively changed to Ne (i-1), u (i). -3), u (i-2), u (i-
1) and replace the previous integral term u ₁ (i) with u ₁ (i−
1), and the latest rotational speed Nen is further substituted into Ne (i). As a result, the current state variable amount [Ne (i) Ne
(I-1) u (i-1) u (i-2) u (i-
3)] is set. And steps 704,70
At 6,708, the same processing as steps 404, 406, 408 in FIG. 8 is performed.

In the open processing, the previous state variable amounts Ne (i), u (i-2), u (i-
1), u (i) is Ne (i-1), u (i-3), u
(I-2), u (i-1), and the latest rotational speed Ne
Substitute n for Ne (i). Thus, the current state variable amount [Ne (i) Ne (i-1) u (i-1) u (i-
2) u (i-3)] is set. Steps 804, 806, and 808 respectively correspond to step 50 in FIG.
The same processing as in 2,506,508 is performed.

Then, in the storage processing, the state variable quantities Ne (i), u (i−i) obtained in the processing shown in FIGS.
2), u (i-1 ), stored in RAM53 in preparation for next processing u a (i) and u ₁ (i) as is.

In the processing of FIG. 7, the initial value of the target rotation speed NF in the F / B processing when the state has shifted from the open state to the F / B state is the rotation number itself at the time when it is determined to start the F / B processing. However, a value obtained by converting a predetermined value to the rotational speed at the time when the F / B process is determined to be started or a value obtained by subtracting the value may be used. Specifically, as shown in FIG.
If it is determined in step 322 that Nen ≦ NFB + NA and the process proceeds to step 323, the correction value NB is set to + α (for example, +50 rpm), and the process proceeds to step 324. Also, LL instead of Nen ≦ NFB + NA:
When proceeding to step 325 assuming that 3 seconds have elapsed since ON,
The correction value NB is set to −α, and the process proceeds to step 324. After the F / B flag is set to 1 in step 324,
In step 328, the calculation result of Nen-NFB + NB is substituted for the lifting amount NFOPEN, and the process proceeds to step 310. First
According to the processing in FIG. 7, when the F / B starts due to a sudden decrease in rotation as after racing, the correction value NB is set so that the lifting amount NFOPEN is increased by a predetermined value. The target rotation speed NF at the start is set to a value higher by the correction value NB than the actual rotation speed at the time of determining that the F / B is started, and is superior to that of the above-described embodiment in suppressing the reduction speed of the rotation speed. It will be. Also, the actual rotation speed Nen remains high but L
L: Lifting amount NFO when F / B starts 3 seconds after ON
Since the initial value of the target rotation speed NF is reduced by reducing PEN by the correction value NB, the reference rotation speed NFB which is the target rotation speed during normal idling after starting F / B than in the above-described embodiment.
To quickly settle down.

Further, in the above embodiment, the description has been made using the apparatus configured to construct the state variable quantity using the value of the past input / output data itself.
An embodiment applied to an apparatus for estimating a state variable quantity by a state observer as disclosed in JP-A-9-7752 will be described. Basically, the entire control program for the ISC valve 44 is the same as that in FIG.
Only the steps 0, 332 and 334 are different, so a portion corresponding to this step will be described with reference to FIGS. 18 to 20. The basic technique is described in the above-mentioned publication, so the description is omitted.

First, in the F / B process shown in FIG. 18, in step 1102, the integral term u stored at the previous calculation timing is obtained.
Calculating the integral term u ₁ of this addition the difference between the target rotation speed NF and the actual rotation speed Nen to _10. In step 1104, the actual rotation speed Nen
Is calculated from the reference set value Na (for example, 650 rpm). In step 1106, the state variable amounts X ₁₀ , X ₂₀ , and X ₃₀ stored at the previous calculation timing, the increment Δu _{0 of} the control value u also stored with respect to the reference set value ua, and the deviation obtained in step 1104. ΔN and the optimal gain (b ₁ , b ₂ , b ₃ , b
₄ ) and (q ₁ , q ₂ , q ₃ , q ₄ ) are weighted and added to determine the current state variable quantities X ₁ , X ₂ , X ₃ , X ₄ . In step 1108, step 1
The current u ₁ obtained in step 102 and the current state variable quantities X ₁ , X ₂ , X ₃ , X ₄ obtained in step 1106 are multiplied by optimal gains K ₁ , K ₂ , K ₃ , K ₄ , K _5. The addition is performed to obtain the current increment Δu. In step 1110, the current control value u is determined from the reference set value ua and the increment Δu. In step 1112, a control signal having a duty ratio corresponding to the determined current control value u is output from the output port 58.
Output to ISC valve 44.

Next, in the open process shown in FIG.
In 02 sets the present control value u to a predetermined value u _0. This predetermined value u ₀ is the same as that of step 502 in the above embodiment. In step 1204, the present increment Δu is obtained from the control value u set in step 1202 and the reference set value ua. In step 1206, a deviation ΔN of the actual rotational speed Nen from the reference set value Na is calculated. In step 1208, the state variable amount X in this open state is prepared in preparation for the next F / B process as in the previous embodiment.
₁ , X ₂ , X ₃ , and X ₄ are obtained by the same processing as the processing in step 1106 in FIG. In step 1210, the present increment Δu obtained in steps 1204 and 1208 and the state variable amounts X ₁ , X ₂ , X ₃ , X
As in the previous embodiments according to _4, inversely calculates u ₁ commensurate with the state at that time. And in step 1212 step 12
A control signal having a duty ratio corresponding to the control value u set in 02 is output from the output port 58 to the ISE valve 44.

Next, in the storage processing shown in FIG. 20, in step 1302, in preparation for the next F / B processing, one of the above-described F / B processing and the open processing is executed, and X ₁ at the present calculation timing is determined. , X ₂ , X ₃ , Δu, u ₁ to X ₁₀ , X ₂₀ , X ₃₀ , Δu ₀ , u
_In step 1304, X ₁₀ , X ₂₀ , X ₃₀ , Δu ₀ , u ₁₀
Is stored in the RAM 53.

That is, also in this embodiment, in the open state, the state variable amounts X ₁ , X ₂ , X ₃ , X ₄ are obtained corresponding to the state at that time, and the state variable amounts and the determined control value u are related. The integral term u ₁ is inversely calculated from the increment Δu. Then, the state variable quantities X ₁ , X ₂ , X ₃ obtained in the open state are replaced with the previous state variable quantities X ₁₀ , X ₂₀ , X ₃₀ currently stored in preparation for the next F / B processing. That is, it is updated and stored. The stores u ₁ that Motoma' inverse operation to place the u ₁₀ stored in the previous calculation timing with regard u _1. In the above embodiment, after the completion of the warm-up, the F / B condition 1
However, this condition may be deleted and the F / B processing may be performed during warm-up. However, to compensate for the lack of air during warm-up, a mechanical air valve is
Preferably, an air valve is juxtaposed with the ISC valve relative to the valve.

Also, considering the movable range and linearity of the ISC valve 44,
Control value (duty ratio) for SC valve 44 is 20% to 80
A second embodiment adapted to the case where the upper and lower limits of% are limited will be described with reference to the flowchart shown in FIG. 21st
In the flowchart shown in the figure, the processing of steps 410 to 413 is inserted between step 404 and step 406 in FIG. 8, and the processing of steps 414 to 416 is inserted between step 406 and step 408. It has been inserted.

First, in step 402, the latest rotational speed Nen is substituted for Ne (i).
₁ Find (i).

In step 410, it is determined whether or not u ₁ (i) obtained in step 404 is larger than a value obtained by adding a predetermined value α to u ₁ (i−1) determined in the previous calculation timing. In step 411, the integral term u obtained in step 404
₁ (i) is u ₁ (i-1) determined at the previous timing
It is determined whether the value is smaller than a value obtained by subtracting the predetermined value β from.
If it is determined in step 410 that u ₁ (i)> u ₁ (i−1) + α, in step 412, u ₁ (i) is set to u ₁ (i).
(I−1) + α is substituted, and in step 411, u ₁ (i) <u ₁
If it is determined that (i-1) -β, step 41
In step 3, u ₁ (i−1) −β is substituted for u ₁ (i). That is, guarding is performed such that the current u ₁ (i) falls within a predetermined range based on the previous u ₁ (i−1).

In the next step 406, the current control value u ₁ (i) is calculated based on the equation (8). Note that the current control value u
₁ Ne (i-1), u used to calculate (i)
(I-1), u (i-2) and u (i-3) have been stored in the storage processing of step 334 in FIG. 7 in preparation for this feedback processing in the previous main processing. N
Based on e (i-1), u (i-1), u (i-2), u (i-3) and Ne (i) obtained by the current processing, Is set. That is, in step 406, the current setting Previously determined Are calculated in a matrix to further determine the current control value u ₁ (i).

Next, in step 414, it is determined whether or not the current control value u ₁ (i) calculated in step 406 is within the range of predetermined upper and lower limits (20 to 80%). In step 415, if the value is out of the range and exceeds the upper limit (80%), the current control value u (i) is set to 80%, and if the value is lower than the lower limit (20%). The current control value u (i) is set to 20%.

After the current control value u (i) is limited in step 415, the process proceeds to step 416. In step 416, in response to the limitation of the control value u (i), the limited control value u (i)
And u ₁ by the following equation so that
(I) is inversely operated. Then, the current control value u set in step 406 or step 415 in step 408
A control signal having a duty ratio corresponding to (i) is output to the output port 58.
To the ISC valve 44.

Therefore, according to the control of the ISC valve 44 as shown in FIG. 21, the final control value u
(I) is controlled within a range between a predetermined upper limit and a lower limit.
Then, while the control value u (i) is limited, u ₁ (i) is inversely calculated so that the limited control value u (i) and the state variable amount satisfy Expression 8.

Therefore, the control value u is set in the direction opposite to the direction in which the restriction is applied.
(I) can be changed sensitively in response to the engine state, so that the valve 44 can be operated responsively. I mean. The problem of the response delay described in the prior art can be sufficiently suppressed.

Further, the initial value (u ₁ (i)) immediately after the control value is shifted from the restricted state to the unrestricted state is determined corresponding to the state variable amount corresponding to the engine state immediately before the transition to the unrestricted state. , The change in u ₁ during the transition can be made smooth, and the control value also changes smoothly, so that the rotation fluctuation at the transition can be suppressed. .

Further, in the embodiment shown in FIG. 21, while the current control value u (i) is restricted in step 415, step 416
And u ₁ (i) is inversely calculated. However, as shown in FIG. 22, while the current control value u (i) is limited in step 415, the current u ₁ (i) is changed in step 417 to u ₁ (i− You may hold it in 1).

Next, the case where the current control value u (i) is limited as described above in the embodiment shown in FIG. 18 will be described with reference to the flowchart shown in FIG. The flowchart shown in FIG. 23 corresponds to steps 1108 and 11 in FIG.
The processing of Steps 1120 to 1122 is inserted between Step 10 and Step 10.

In step 1120, the present increment Δu (i) produced in step 1108 as described above is set to a predetermined upper / lower limit (Δu _min
~ Δu _max ). Then, in step 1121, the current increment Δu (i) is out of the range,
If it exceeds the upper limit value Δu _man , the current increment Δu (i) is set to the upper limit value Δu _max . If the difference is smaller than the lower limit Δu _min , the current increment Δu (i) is set to the lower limit Δu _min . Then, in step 1122, u
₁ Invert (i) by the following equation.

u ₁ (i) = − (Δu (i) + K ₂ · X ₁ (i) + K ₃ · X
In _{2 (i) + K 4 ·} X 3 (i) + K 5 · X 4 (i)) / K 1 step 1110, the current increment Δu set in step 1108 or step 1121 (i) and the reference set value ua The current control value u is set according to.

Next, an embodiment in which the present invention is applied to an idle speed control device using a step motor control valve (step motor control valve) as an air control valve will be described.

The contents of processing executed by the CPU 51 of the electronic
This will be described with reference to the flowchart shown in FIG.

The flowchart of FIG. 24 is a control program of the step-type control valve, and is similar to the above-described embodiment of the ISC valve.
When the process is started for a predetermined time (for example, every 100 msec) with the G switch closed, step 200
In step 1, it is determined whether the operating state of the engine 10 meets the conditions for executing the idle speed control. As the feedback condition, as shown in the flowchart of FIG. 7, a predetermined time (for example, 3 seconds) after the start is completed.
ec) It has elapsed, the throttle valve 31 has been fully closed, and the warm-up has been completed in some cases. When all the conditions are satisfied, the routine proceeds to step 2004, where the engine speed reaches the target engine speed. The feedback control is performed so that the number becomes a number.

On the other hand, if it is determined in step 2001 that the feedback condition is not satisfied, an open process from step 2002 is performed.

In this open process first sets the current control value u (i) to a predetermined value u ₀ at step. The predetermined value u ₀ may be an arbitrary constant value such as 100%, 0% or 50% as a duty ratio, or may be a value determined according to a detection parameter such as a cooling water temperature Thw.

Next, in step 2003, the detected rotation speed Nen is substituted for Ne (i). Then go to step 2016.

In step 2004, the cooling water temperature Thw, the intake air temperature Tam,
The target speed NF is determined based on the on / off state of the air conditioner, the range position of the automatic transmission, and the like.

In step 2005, the latest detected rotation speed Nen is set to Ne (i)
In step 2006, the integral value u ₁ (i) is updated by equation (9) based on the deviation between the target rotation speed NF determined in step 2004 and Ne (i) determined in step 2005. .

In the next step 2007, the current control value u (i) is calculated based on equation (8). Note that the current control value u
Ne (i-1), u (i) used to calculate (i)
-1), u (i-2) and u (i-3) are stored in preparation for the feedback processing in step 2022 described later in the previous processing, and these Ne (i-1), u
(I-1), u (i-2), u (i-3) and Ne (i) obtained in the current processing, Is set.

In other words, in step 2007, the current Previously determined Are calculated in a matrix, and the current control value u (i) is determined by further adding the current u ₁ (i).

Next, in step 2008, it is determined whether or not the current control value u (i) calculated in step 2007 is within the range of predetermined upper and lower limits (20 to 80%). In step 2009, if the value is out of the range and exceeds the upper limit (80%), the current control value u (i) is set to 80%, and if it is lower than the lower limit (20%). The current control value u (i) is set to 20%.

After the current control value u (i) is limited in step 2009, the process proceeds to step 2016.

On the other hand, if the current control value u (i) is within the above upper and lower limits in step 2008, the process proceeds to step 2010. In step 2010, it is determined whether or not the absolute value of the difference between the detected rotation speed Ne (i) and the target rotation speed NF is equal to or less than a predetermined value (for example, 25 rpm in the present embodiment). If the absolute value of the deviation is equal to or smaller than the predetermined value, the flag FSTA is set in step 2011 (FSTA = 1).
It is detected whether or not it has been performed. Here, the flag FSTA is set when the rotation speed is in the steady state near the target rotation speed during the feedback control. Steps
If the flag FSTA is set in 2100, that is, if the flag FSTA is in a steady state, the process proceeds to step 2015.

If the flag FSTA is not set in step 2011, the counter C is counted up (C ← C + 1) in step 2012. Here, the counter C measures the elapsed time after the absolute value of the above-described deviation has become equal to or less than a predetermined value. In the following step 2013, a predetermined time (for example, 5 seconds in this embodiment) has elapsed after the counter C has become equal to or more than a predetermined value (for example, 50 in this embodiment), that is, the absolute value of the above-described deviation has become equal to or smaller than the predetermined value. It is determined whether or not. If the predetermined time has not elapsed, the process proceeds to step 2019.

On the other hand, if the predetermined time has elapsed, it is determined that the vehicle is in a steady state, and the flag FSTA is set in step 2104 (FSTA
← 1) and proceed to step 2015.

If the vehicle is in the steady state, the current control value u (i) is set to the steady state control value uAV (i) in step 2015, and the process proceeds to step 2016.

In step 2016, in step 2009, the current control value u
When (i) is restricted, step in the open process
When the current control value u (i) is set to the predetermined value u (i) in 2002, and in the steady state, the current control value u (i) is set to the steady state control value uAV (i) in step 2015. Is performed, that is, the current control value u (i) is Is set to a value unrelated to the current control value u
(I) and And u ₁ (i) are inversely calculated by the following equation so that satisfies equation (8).

u ₁ (i) ← u (i) −K ₁ · Ne (i) −K ₂ · Ne (i−
_{1) -K 3 · u (i} -1) -K 4 · u (i-2) -K 5 · u (i-
3) Then, proceed to Step 2019. In step 2019, a smoothing process is performed on the current control value u (i) by the following equation to calculate a steady-state control value uAV (i).

uAV (i) = {7 × uAV (i-1) + u (i)} / 8 where uAV (i-1) is a control value in a steady state calculated at the previous control timing.

Subsequently, in step 202, step 2002, step 200
7. A control signal having a duty ratio corresponding to the current control value u (i) set in any of Steps 2009 and 2015 is output from the output port 58 to the step motor control valve.

Then, in step 2021, Ne (i), u (i-2), u (i-1), u (i), u set as described above are set.
₁ (i) and uAV (i) are Ne (i-1) and u (i-
3), u (i-2 ), u (i-1), u 1 (i-1), uAV
Substitute (i). In the following step 2022, step 2021
Ne (i-1), u (i-3), u (i-2), u
_{(I-1), u 1} (i-1), stores uAV (i) to RAM 53,
This processing ends.

In step 2019 described above, the control value uAV in the steady state
(I) is set by smoothing the current control value u (i), and the average value of the control value u (i) is set to the steady-state control value uAV (i) as in the following equation. You may make it.

uAV (i) = ｛u (i) + u (i−1) + u (i−
2) A steady state (for example, a state in which 5 seconds or more have elapsed since the deviation between the target rotation speed NF and the rotation speed became 25 rpm or less in the present embodiment) by the processing of + u (i-3)｝ / 4 or more. )
, The current control value u (i) is replaced with the steady-state control value
Set to uAV (i). This steady state control value uAV (i)
Is not updated if it is in a steady state. Therefore, a change in the operation amount of the step motor control valve can be suppressed, and the durability thereof can be improved.

Also, the control value u (i) of this time is the control value uAV of the steady state.
While this is set to (i), the current control value u (i) The integral term u ₁ (i) is inversely calculated such that satisfies Expression (8). Therefore, it is possible to suppress the rotation fluctuation at the time of transition from the steady state to the unsteady state.

As described above, especially in FIG. 7, the state variable amount is always set even in the open loop. Therefore, even if the current control value shifts from a state in which the current control value is set irrespective of the current state variable amount to a state in which the current control value is set in accordance with the current state variable amount, the current state variable amount Has already been set, the adjusting means can be operated with good response, and there is an excellent effect that the convergence to the target rotational speed is improved.

Next, a description will be given of an embodiment improved in order to eliminate a problem that, when a load such as a D range is input, the control value is corrected in a feedford manner and the rotational speed is higher than the target rotational speed. I do.

In the present embodiment, the control value (duty ratio) for the ISC valve 44 is limited between the upper and lower limits of 0% to 100% in accordance with the movable range of the ISC valve 44. The state variable X
Is limited by a predetermined upper and lower limit also _{1 (i) ~X 5 (i} ) and the integral term u ₁ (i).

Further, in the present embodiment, the feedforward control is also performed according to the state of the load on the engine 10 such as the air conditioner or the automatic transmission described above, and the state of the rotational speed Ne obtained based on the output from the rotational speed sensor 30. In particular, in this embodiment, the feedforward processing corresponding to the control value, the state variable amount, and the integral term whose upper and lower limits are respectively set as described above is executed.

The processing executed by the CPU 51 of the electronic control unit 20 will be described below with reference to FIGS. 25 and 26.

The flowchart of FIG. 25 is a control program of the ISC valve 44, which is executed by interruption every predetermined time (for example, every 100 msec) while an IG switch (not shown) is closed.

First, when the process is started by an interrupt, it is determined in step 3302 whether the operating state of the engine 10 meets the condition for executing the feedback control of the idle speed. The feedback conditions include that a predetermined time (for example, 3 seconds) has elapsed after the start is completed, the throttle valve 31
May be fully closed, and after warm-up is complete,
When all the conditions are satisfied, the process proceeds to step 3304.

In step 3304, it is determined whether the neutral switch 64 is on, that is, whether the automatic transmission is in a neutral range (neutral or parking). Low, second, drive, reverse) and the switch 64 is off, the process proceeds to step 3308.

Step 3306 sets the reference rotation speed NFB to 700 rpm,
In the next step 3310, 5 rpm is added to the previously set target rotation speed NF. Next in step 3312
The target rotation speed NF obtained in 3310 is compared with the reference rotation speed NFB. If NF> NFB, the reference rotation speed NFB is set as the target rotation speed NF in step 3314.

On the other hand, in step 3308, the reference rotation speed NFB is set to 600 rpm, and in next step 3316, 5 rpm is subtracted from the previously set target rotation speed NF. Next, in step 3318, the target rotation speed NF obtained in step 3316 is compared with the reference rotation speed NFB, and if NF <NFB, the reference rotation speed NFB is set as the target rotation speed NF in step 3320.

In other words, in steps 3304 to 3320, the basic level of the target rotation speed is switched between when the automatic transmission is in the neutral range and when the automatic transmission is in the traveling range, and immediately after switching between the neutral range and the traveling range, immediately after the switching. Instead of setting the target rotation speed according to the range, the target rotation speed is gradually changed to the target rotation speed according to the switched range over time. Specifically, immediately after switching from the neutral range to the running range, the target rotation speed NF is reduced from 700 rpm by 5 rpm every 100 msec to 600 rpm.
Conversely, immediately after the traveling range is switched to the neutral range, the target rotational speed NF is increased from 600 rpm to 700 at 5 rpm every 100 msec.
Increased to rpm.

As described above, the reason why the basic level (reference speed NFB) of the target speed is lower in the traveling range than in the neutral range is to suppress the creep phenomenon.

Next, in step 3322 and step 3324, respectively, the previous state of the neutral switch 64 is checked, and when it is determined that the state of the neutral switch 64 is different between the previous time and the current time, and that the switching of the range of the automatic transmission has occurred, integral term is stored in the RAM53 as described above u ₁ a (i-1) to modify only a feed forward manner once in step 3326,3328. Specifically, in step 3326, the automatic transmission is switched from the running range to the neutral range, and the load on the engine 10 is reduced. Therefore, the integral term u ₁ (i−1) is corrected by β ₁ only once after the switching. Since the load on the engine 10 increases in step 3328, the integral term u ₁ (i-1) is corrected by α ₁ only once after the switching.

Next, in steps 3330 and 3332, it is determined whether the air conditioner switch 62 has been switched between the previous time and the current time, that is, whether the vehicle air conditioner has been turned on or off. Then, in step 3334 if the air conditioner is determined to have been switched on from the off state, since the load on the engine 10 increases, alpha ₂ only integral term u ₁ (i-
1) is increased only once after the switching, and if it is determined that the air conditioner has been switched off from the on state, the integral term u ₁ (i−1) is switched by β _{2 in} step 3326 in step 3326. Is corrected only once.

Next, in steps 3338 and 3340, it is determined whether the fluctuation of the rotation speed obtained by subtracting the detection rotation speed Nen at the current processing timing from the detected rotation speed Neo at the previous processing timing, particularly the descending amount is larger than a predetermined value A, increases corrects the integral term u _{1 (i-1)} only alpha ₃ is greater.

Next, in steps 3342 and 3344, the detected rotation speed Nen is set to the upper and lower limit values Ne max (for example, 1600 rpm) and Ne min as the idle rotation speed.
(For example, 400 rpm), and determines the lower limit value Ne.
If it is determined that the detected rotation speed Nen is below than min, integral term u ₁ a (i-1) increased corrected by alpha ₄ at step 3346, when from the upper limit value Nemax is detected rotation speed Nen exceeds reversed If it is determined, in step 3348, the integral term u ₁ (i−
1) is reduced and corrected by β ₄ .

Next, in step 3350, the lower limit Ne min is substituted for Ne (i) used in the calculation described later.
In 3352, the upper limit Ne max is substituted for Ne (i), and in step 3354, the detected rotational speed Nen is substituted for Ne (i). That is, in steps 3350, 3352, and 3354, Ne (i) is limited to a range between predetermined upper and lower limits (Ne min to Ne max).

In the following step 3356, determined by the above processing,
Or modified NF, Ne (i), and u _{1 (i-1)}
Is used to calculate the current integral term u ₁ (i) based on the above equation (9). Then, in the next steps 3358 and 3360, it is determined whether or not the calculated integral term u ₁ (i) is within the range of predetermined upper and lower limits (0% to 100%). At 3362, the integral term u (i) is set to 100%, and if it is below the lower limit (0%), at step 3364, the integral term u ₁ (i) is set to 0%. ₁
(I) The upper and lower limits are the limit value u obtained by the above equation (8).
According to (i), the range is set so that the ISC valve 44 can be actually moved.

In the next step 3366, the current control value u (i) is calculated based on the equation (8). The current control value u (i)
Ne (i-1), u (i-
1), u (i-2), u (i-3) are stored in preparation for this feedback processing in the previous processing, and these Ne (i-1), u (i-1) , u (i-2), u
(I-3) and Ne (i) obtained in the current processing, Is set.

In other words, in step 3366, the current Previously determined Are calculated in a matrix, and the current integral term u ₁ (i) is added to determine the current control value u ₁ (i). The above Ne (i) and Ne (i-1) in step 3342 and 33
44, 3350, 3352, and 3354 limit the range to the upper and lower limits.

Next, in steps 3368 and 3370, the current control value u (i) calculated in step 3366 is set to a predetermined upper / lower limit (0% to 100%).
Is determined, the control value u (i) is set to 100% in step 3372 if it exceeds the upper limit (100%), and if it is below the lower limit (0%) in step 3372. Steps
At 3374, the current control value u (i) is set to 0%. Note that the upper and lower limits of the control value u (i) are set so as to be in a range in which the ISC valve 44 can be actually moved. By limiting the control value u (i) in a predetermined range in this way, U (i-1), u (i-2), and u (i-3) are also limited to predetermined upper and lower limits (0% to 100%).

In step 3376, a control signal having a duty ratio corresponding to the current control value u (i) determined as described above is output from the output port 58 to the ISC valve 44.

If it is determined in step 3302 that the feedback condition is not satisfied, the process proceeds to step 390 open processing.

FIG. 4 shows the contents of the open processing. This open process first sets the current control value u (i) to a predetermined value u ₀ at step 402. The predetermined value u ₀ may be an arbitrary constant value such as 100%, 0% or 50% as a duty ratio.
Further, the value may be a value determined according to a detection parameter such as the cooling water temperature Thw. Then steps 404,406,408,410,
In step 412, as in steps 342, 344, 350, and 354 described above, the upper and lower limits Ne max, Ne min
When Nen> Nemax, Nemax is substituted for Ne (i), when Nen <Nemin, Nemin is substituted for Ne (i), and when Nemin ≦ Nen ≦ Nemax, Ne is substituted. (I) is substituted for Nen. Next, in step 414, Ne (i) set in steps 408, 410, and 412 and Ne (i-
1), u (i-1), u (i-2), u (i-3) and steps
Based on the current control value u (i) set in 402, the control value u (i) set this time and the integral term u ₁ (i) that matches the current state variable amount are calculated inversely based on equation (8). I do.

Note that the state variable amount at the time of this open processing is Ne (i) set in steps 3408, 3410, 3412 and Ne (i-1), u (i-1), u (i-) stored in the RAM 53.
2) From u (i-3) to [Ne (i) Ne (i-1) u
(I-1) u (i-2) u (i-3)].

In subsequent steps 3416, 3418, 3420, and 3422, the integral term u ₁ (i) obtained in step 3414 is 0% to 100%.
Determines it is within bounds, if exceeds 100% integral term u ₁ (i) is set to 100%, the integral term u ₁ if you are below 0% in the reverse ( Set i) to 0%. When the above processing is completed, the above step 3376
Proceed to.

After executing the processing in step 3376, in step 3378, the feedback processing of steps 3304 to 3374 and the step
Ne (i), u (i-2), and u (i-1) in the state variable amounts set in any of the open processing of 3390 are Ne (i-1), u (i-3), u (i-2), and the current control values u (i) and u ₁ (i) determined by the feedback processing or the open processing are respectively set to u
(I-1), is substituted for u _{1 (i-1).}

Next, at step 3380, the Ne determined at step 3378 above
(I-1), u ( i-3), u (i-2), u (i-1), u 1
(I-1) is stored in the RAM 53.

That is, in the storage process, Ne (i), u (i-2), u (i-1) used in the feedback process and the open process and the control value u (i) determined in each process are used in the next process. The stored state variable amount is updated and stored in preparation for the feedback process and the process of the inverse operation of the integral term in the next open process. Further, the integral term u ₁ (i) determined in the feedback processing is also stored in preparation for the next feedback processing. Further, the integral term u ₁ (i) calculated by the open processing is also stored as an initial value when the integral term is calculated by Expression (9) in the next feedback processing. Moreover, in this embodiment, the data is stored after being changed to the form used in the processing at the next calculation timing (step 3378).

Therefore, according to the control program of the ISC valve 44,
The integral term u ₁ (i stored in correspondence with the switching of the load state on the engine 10 such as an air conditioner or an automatic transmission.
-1) is corrected once only in a feedforward manner, and the corrected integral term u ₁ (i-1) is used for the current integral term u ₁ (i-1).
₁ (i) is determined and the current integral term u ₁ (i) is restricted to be within the upper and lower limits, so that the integral term u ₁ (i) is not affected by the feedback forward correction. , Since it is possible to fully vary within the upper and lower limits, the control value u
₁ (i) can also be fully varied, and the control value u (i) can be reduced despite the detected rotation speed being higher than the target rotation speed due to the operation state of the load, as in the above-described conventional technology. The risk of causing the problem that the ISC valve 44 cannot be closed without any problem can be eliminated.

Moreover, since the modified feedforward immediately integral term u _{1 (i-1)} if the idle speed of descent occurs, it can be sufficiently prevented stalling even when the load of the engine 10 is rapidly increased by some disturbance . Furthermore, when the idle speed reaches the predetermined upper and lower limits, the integral term u ₁ (i−1) is corrected in a feedforward manner in a direction falling within the upper and lower limits, so that the load on the engine 10 is suddenly increased. Increase
Or, even if the idle speed decreases greatly from the target speed and reaches the upper and lower limits,
Since the opening of the ISC valve 44 is increased or decreased in response to the feedforward correction, the ISC valve 44 is immediately returned to the target rotational speed side. In addition, since the feedback process is performed at the same time, the idle speed can be smoothly returned to the target speed.

In the above process, the state variable amount is updated and the integral term is prepared in preparation for the next feedback process at the time of the open process.
Since the inverse operation of u ₁ (i) is performed, the idle speed immediately after the feedback is entered from the open state is set very smoothly and quickly to the target value.

Also, in the idle speed control using the above-described modern control which is more excellent in responsiveness, when the target speed is rapidly changed, overshoot and undershoot increase, causing large rotation fluctuation. When switching between the neutral range and the traveling range of the automatic transmission, the target speed corresponding to the range specified to be switched to removal according to the passage of time from the switching point is not changed immediately to the target speed for the neutral range or the traveling range. Since the number of rotations is changed, the large rotation fluctuation due to the overshoot and undershoot described above is eliminated. Moreover, in the above-described processing, the target rotation speed is reduced when the vehicle is set to the traveling range in order to prevent creep, and the target rotation speed is reduced when the load is increased. Although the possibility of causing engine stall is extremely large, as described above, this is particularly effective in the case where the target rotational speed must be reduced despite the increased load, and the responsiveness is high. This is an indispensable technology in idle speed control using modern control.

By the way, in the processing of FIG. 25, the stored integral term u ₁ (i−1) is corrected by feedforward, and the corrected integral term u ₁ (i−1) is further modified by NF, Ne (i) and Although this is subjected to cumulative processing in response to the difference between the integral term u ₁ to obtain a (i), those feedforward Fixed as modified feedforward as the current integral term u ₁ (i) is NF and Ne The accumulation process according to the difference from (i) may not be performed.

In the above processing, the upper and lower limit values for the detected rotation speed Nen for feedforward correction of the integral term u ₁ (i−1) in steps 3342 to 3354 and the upper and lower limit values of Ne (i) used as the state variable amount Have the same values as Ne max and Ne min, but the upper and lower limits may be different values.

In the above process, when the range of the automatic transmission is switched, when the air conditioner is switched, when the rotation speed drops rapidly, or when the rotation speed reaches the upper and lower limits, the predetermined amount (α ₁ , α ₂ , α ₃ , α ₄ , β ₁ , β ₂ , β ₄ )
The integral term u ₁ (i) was corrected by directly adjusting u ₁ (i−1). However, as shown in FIG. The integral term u ₁ (i) may be modified by adding or subtracting the number NF to form a tentative target rotational speed NF ′. In FIG. 5, in detail, any of the steps 3322 and 3324 and the steps 3330, 3332, 3338, 3
Step 336 via 342, step 3344, step 3354
If you go to 5. The same processing as described above is executed, and conversely, if any of the above changes occurs, step 3326 ',
In any of step 3328 ', step 3334', step 3336 ', step 3340', step 3346 ', and step 3348', the provisional target rotation speed NF 'is set.
65 'defines the current integral term u ₁ (i).

Step 3326 ', step 3328', step 33
34 ', step 3336', step after step 3340 '
In the case of proceeding to 3365 ', Ne (i) is set in step 3341 by the setting process of Ne (i), that is, the same process as steps 3342, 3344, 3350, 3352, and 3354, and then the process proceeds to step 3365'.

Further, in the above embodiment, the description has been made using the apparatus configured to construct the state variable quantity using the value of the past input / output data itself, but the state observation as shown in JP-A-59-46353 is described. An embodiment in which the present invention is applied to an apparatus for estimating a state variable quantity by a vessel will be described with reference to FIG. In addition,
Basically, the overall control program for the ISC valve 44 is the same as that of FIG. The basic technology is described in the above-mentioned publication, and the description is omitted.

In FIG. 28, steps 3302 to 3348
It is the same as the figure. When any of the processes in steps 3344, 3346, and 3348 is completed, the integral term u ₁ stored in the previous calculation timing in step 3602 and corrected as necessary by the processes in steps 3304 to 3348 described above. (I-1),
And the integral term u ₁ (i) is determined based on the deviation between the target rotational speed NF and the detected rotational speed Nen. Next, at step 3604, it is determined whether or not the obtained integral term u ₁ (i) is within the range of upper and lower limits (−u _{a to} 100−u _a ).
Limited by 100-u _a or the lower limit -u _a. Step 3608
Calculates a deviation ΔN of the detected rotation speed Nen from a reference set value Na (for example, 650 rpm). In step 3616, the state variable amounts X ₁ (i−1), X stored at the previous operation timing
_{2 (i-1), X} 3 (i-1) and, likewise increment for reference set value u _a memory once was the control value u (i) Δu (i-
1) and the deviation ΔN obtained in step 3608 are weighted and added by the optimal gains (b ₁ , b ₂ , b ₃ , b ₄ ) and (q ₁ , q ₂ , q ₃ , q ₄ ) to change the state. The quantities X ₁ (i), X ₂ (i), X ₃ (i), and X ₄ (i) are obtained. In steps 3612 and 3614, it is determined whether each of the values X ₁ (i) to X ₄ (i) constituting the state variable amount obtained in step 3610 is within predetermined upper and lower limits. If there is, it is limited by the upper limit or lower limit. In step 3616, the integral term u ₁ obtained this time through the processing in steps 3602 to 3606
The optimum gains K ₁ and K ₂ are obtained for the current state variable quantities X ₁ (i), X ₂ (i), X ₃ (i) and X ₄ (i) obtained through (i) and the processing of steps 3610 to 3614. , K ₃ , K ₄ , K ₅ and add the current
Find (i). In step 3618 the control value u (i) is determined from the reference set value u _a and incremental Delta] u (i). Step 36
At 20,3622, it is determined whether or not the control value u (i) obtained at step 3618 is within a predetermined range of upper and lower limits (0 to 100). ) To determine the current control value u (i). And step
At 642, a control signal having a duty ratio according to the determined current control value u (i) is output from the output port 58 to the ISC valve 44.

Next, in the open process performed when it is determined in step 3302 that the feedback condition is not satisfied, in step 3624
In setting the current control value u (i) to a predetermined value u _0. This predetermined value u ₀ is the same as that in step 3402 in the above embodiment. Step from the control value set in step 3624 in 3626 u (i) and the reference set values u _a of this increment Δu
Find (i). In step 3628, a deviation ΔN of the detected rotational speed Nen from the reference set value Na is calculated. In step 3630, the state variables X ₁ (i), X ₂ (i), and X _{3 in} this open state are prepared in preparation for the next feedback processing as in the above embodiment.
(I), X ₄ (i) are obtained by the same processing as the processing in step 3610. In steps 3632 and 3634, steps 3612 and 3614
Each value X ₁ (i) to X ₄ constituting the state variable quantity in the same process as
(I) is limited to the upper and lower limits. In step 3636, the present increment Δu obtained through each step of the above open processing
(I) and the state variable quantities X ₁ (i), X ₂ (i), X
₃ (i) and X ₄ (i), as in the previous embodiment, the integral term
Inverse operation of u ₁ (i). In steps 3638 and 3640, the integral term u ₁ (i) obtained in the same manner as in steps 3604 and 3606 is limited to the upper and lower limits. And in step 3642, step 36
A control signal having a duty ratio corresponding to the control value u ₁ (i) set at 24 is output from the output port 58 to the ISC valve 44.

Next, in preparation for the next feedback process, at step 3644, X at the current calculation timing determined by execution of either the above-described feedback process or open process is determined.
_{1 (i), X 2 (} i), X 3 (i), Δu (i), u 1 (i) , respectively _{X 1 (i-1),} X 2 (i-1), X 3 (i- 1), Δu
(I-1), u ₁ (i-1).
X ₁ (i−1), X ₂ (i−1), X ₃ (i−1), Δu (i−
1), u ₁ (i−1) is stored in the RAM 53, the current calculation and control processing is ended, and the processing shifts to another processing.

In the above-described embodiment, one of the F / B conditions is set after completion of warm-up. However, this condition may be deleted, and F / B processing may be performed during warm-up. However, it is preferable to provide a mechanical air valve in parallel with the ISC valve 44 in order to compensate for an insufficient amount of air during warm-up.

Further, in each of the above-described embodiments, the calculation is performed at regular intervals, but the calculation may be performed each time a predetermined rotation angle is detected.

Further, in the above-described embodiment, the target rotation speed NF is changed because the target rotation speed NF is gradually changed at the time of switching the range of the automatic transmission by the processing of steps 3304 to 3320 in FIG. During this period, the detected rotation speed Nen is deviated from the target rotation speed NF. Therefore, the process of step 3356 in FIG. 25 is performed by the following equation.

U ₁ (i) = 2 · u ₁ (i−1) −u ₁ (i−2) + Ka ·
(NF-Ne (i)) Further, in step 3378, u ₁ (i
-2) ← u ₁ (i-1) is added, and step
At 3380, u ₁ (i-2) is also stored. In this way, even if the target rotational speed NF changes as described above, the detected rotational speed Nen follows the target rotational speed NF without deviation or with a very small deviation.
Note that step 3365 'in FIG. 27 and step 3602 in FIG.
The same applies to.

As described above, according to the present embodiment, when the normal change of the load is detected by the state change detecting unit, the integral value calculated by the integral value calculating unit at that time is calculated by the integral value correcting unit.
Since the integral value is corrected in a feedforward manner before being limited to a predetermined range by the integral value limiting means in accordance with the state change, the integral value can be fully varied within the predetermined range, and the feedforward correction according to the load disturbance is performed. However, the control value can be fully varied within a predetermined range. Therefore, even if the detected rotational speed is higher than the target rotational speed and the final control value is minimized as in the prior art described above, a correction amount is added if a load disturbance is present at that time, and the final control value And the idle speed control valve is kept open by the predetermined amount, and the detected rotation speed cannot be converged to the target rotation speed. Can be solved, and there is an excellent effect that the detected rotation speed can be accurately converged to the target rotation speed.

INDUSTRIAL APPLICABILITY As described in detail above, according to the present invention, in an idle speed control device for an internal combustion engine, an actual speed is quickly converged to a target speed even when there is a disturbance, according to the present invention. It becomes possible.

Continued on the front page (72) Inventor Shigenori Isomura 3-617 Harazaki-cho, Kariya-shi, Aichi (72) Inventor Toshio Kondo 1-1-12-6, Suehiro-cho, Kariya-shi, Aichi (56) References JP-A-64-8336 JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) F02D 41/00-45/00 395

Claims (17)

(57) [Claims] 1. A rotational speed detecting means for detecting a rotational speed of an internal combustion engine, a rotational speed adjusting means for adjusting a rotational speed of the internal combustion engine, and a target rotational speed detected by the rotational speed detecting device is a desired target rotational speed. A control value for controlling the rotation speed adjusting means so as to match the number of the rotation speed control means at predetermined intervals, and a control means for outputting a control signal according to the control value, a digital control device comprising: The control unit sets a state variable amount according to the detected rotation speed, a state variable amount set at a previous calculation timing, and the control value corresponding to the control signal output to the rotation speed adjustment unit. A first control value setting unit for setting a current control value according to the state variable amount; a state variable amount set at the previous calculation timing and a previous calculation timing; The output control signal A second control value setting means for setting a state variable amount in accordance with the control value corresponding to the above, and setting a current control value to a value irrelevant to the state variable amount; And a selecting means for selecting one of the first control value setting means and the second control value setting means. 2. The method according to claim 1, wherein the first control value setting means is configured to determine a current integral term based on a value corresponding to a deviation between the detected rotational speed and the target rotational speed and an integral term calculated at a previous calculation timing. 2. A digital system according to claim 1, further comprising: an integral term calculating means for calculating a current control variable, and a control value calculating means for calculating a current control value in accordance with the current state variable amount and the current integral term. Control device. 3. The integral term setting means, wherein when the second control value setting means is selected by the selection means, the integral term of the integral term is set according to the current control value and the current state variable amount. 3. The digital control device according to claim 2, further comprising an integral initial value setting means for setting an initial value. 4. The digital apparatus according to claim 3, wherein said selecting means includes first selecting means for selecting said first control value setting means when said internal combustion engine is in an idle state. Control device. 5. The control unit according to claim 1, wherein the control unit sets the detected rotation speed when the internal combustion engine shifts from a non-idle state to an idle state as an initial value of the target rotation speed. 5. The digital control device according to claim 4, further comprising: a target rotation speed setting unit configured to change a target rotation speed at a predetermined rate from the target rotation speed initial value to a predetermined rotation speed. 6. The apparatus according to claim 5, wherein said selecting means includes second selecting means for selecting said second control value setting means when said control value is out of a predetermined range. Digital controller. 7. When the internal combustion engine is controlled by a control value set by the first control value setting means, the selection means responds to a deviation between the detected rotation speed and the target rotation speed. 7. The digital control device according to claim 6, further comprising third selection means for selecting the second control value setting means when the value is equal to or less than a predetermined value. 8. When the second control value setting means is selected by the third selection means, the second control value setting means sets the current control value at a previous calculation timing. 8. The digital control device according to claim 7, further comprising resetting means for resetting a value according to the control value. 9. The apparatus according to claim 8, wherein said resetting means comprises control value correcting means for correcting the control value set by said first control value setting means by a predetermined filter processing. Digital controller. 10. A rotational speed detecting means for detecting the rotational speed of the internal combustion engine, an adjusting means provided in the engine for adjusting the rotational speed of the internal combustion engine, and a control value for controlling the adjusting means is set to a predetermined value. And a control means for calculating a control signal in accordance with the control value, wherein the control means outputs a control signal according to the control value. Discriminating means for discriminating that the internal combustion engine is in a state where feedback control is performed so as to coincide with the rotational speed detected by the revolution number detecting means when the discriminating means determines that feedback control is performed. The current state according to the state variable amount stored at the previous calculation timing and the control value corresponding to the control signal output to the adjustment means or a value related to the control value. The quantity is determined, and the control value of this time is feedback-calculated from the integrated value obtained by integrating the deviation between the rotation speed detected by the rotation speed detection means and the target rotation speed and the determined state variable amount, and the determined control value is determined. First operation processing means for storing and storing the state variable amount in preparation for the next feedback operation; and setting the current control value to a predetermined value when the determination means determines that the state is not in the state of performing feedback control. The updated state variable amount is updated in accordance with the set control value or the value related to the control value, the rotation speed detected by the rotation speed detection means, and the stored state variable amount, and A second arithmetic processing means for performing storage processing in preparation for the feedback arithmetic operation; and the control signal according to the control value obtained by one of the first arithmetic processing means and the second arithmetic processing means. Digital control apparatus characterized by comprising an output means for outputting. 11. A rotational speed detecting means for detecting the rotational speed of the internal combustion engine, a rotational speed adjusting means for adjusting the rotational speed of the internal combustion engine, and a rotational speed detected by the detecting means coincides with a desired target rotational speed. A control means for calculating a control value for controlling the adjusting means at predetermined intervals, and outputting a control signal according to the control value, the control means comprising: State variable amount setting means for setting a state variable amount according to a rotation speed and the control value corresponding to the control signal or a value related to the control value which is output to the adjusting means and the setting means; State variable amount control means for judging whether the state variable amount set in the above is within a predetermined range, and restricting the state variable amount to a predetermined range when the state variable amount is out of the range, the detected rotation speed and the target rotation speed Find the integral of the deviation of Value calculating means, determining whether the integrated value calculated by the calculating means is within a predetermined range, and limiting the integrated value to a predetermined range when the integrated value is out of the range; Control value calculating means for calculating a control value using the state variable amount obtained via the quantity limiting means and the integral value obtained via the integral value limiting means; control calculated by the calculating means Control value limiting means for determining whether the value is within a predetermined range, and limiting the control value to a predetermined range when the control value is out of the range; and state change detecting means for detecting that a load state on the internal combustion engine has changed. A state change detecting means for detecting that a load state on the internal combustion engine has changed, and when the state change is detected by the state change detecting means, the integrated value calculated by the calculating means is detected. Respond to state changes Digital control apparatus characterized by comprising an integration value correction means for correcting Te. 12. The vehicle according to claim 11, wherein said internal combustion engine is mounted on said transmission-equipped vehicle, and said state change detecting means detects a switch between a traveling range and a neutral range of the automatic transmission. Digital control device. 13. The digital control device according to claim 12, wherein said first control means includes a range discriminating means for discriminating between a traveling range and a neutral range of said automatic transmission, and a range discriminated by said range discriminating means. A target rotation speed setting means for selecting the target rotation speed and gradually changing the target rotation speed to a target rotation speed according to the changed range after the range is switched. . 14. The integral value calculating means uses a deviation between the detected rotational speed and the target rotational speed, an integral value obtained at a previous operation timing, and an integral value obtained at a previous operation timing. 14. The calculation according to claim 13,
An idle speed control device for an internal combustion engine according to claim 1. 15. A detecting means for detecting an operation state of a controlled object, an actuator for driving a driving member for adjusting an operation of the controlled object, and an actuator in accordance with the operating state detected by the detecting means. A control means for calculating control data for control at predetermined intervals and outputting a control signal in accordance with the control data, wherein the control means includes a dynamic model constructed in advance. Determining means for determining a state variable amount based on the control data obtained at the calculation timing and the operating state checked by the detecting means, based on the control timing; and Output means for outputting the control signal according to the state variable amount.The dynamic model is created after the control signal is input to the actuator. It is constructed based on a first dynamic model in which the dead time until the state starts to change starts is identified by a discrete system and a second dynamic model in which the part after the dead time is identified by a discrete system. A digital control device characterized by the above-mentioned. 16. A detecting means for detecting the number of revolutions of the internal combustion engine, an actuator for adjusting the number of revolutions, and control data for controlling the actuator according to the number of revolutions detected by the detecting means is provided at a predetermined cycle. And a control means for calculating each time and outputting a control signal corresponding to the control data, wherein the control means obtains at the calculation timing based on a dynamic model constructed in advance. Determining means for determining a state variable amount according to the control data and the number of revolutions detected by the detecting means; and the control signal according to the state variable amount so that the number of revolutions becomes a desired number of revolutions. The dynamic model comprises a discrete time delay from when the control signal is input to the actuator until the rotational speed starts to change. In digital controller, characterized in that those built on the first dynamic model and the subsequent dead time portion identified in the second dynamic model identified by the discrete system. 17. The control device according to claim 1, wherein the predetermined cycle is
17. The digital control device according to claim 16, wherein the dead time is 1 / N (N is an arbitrary integer).