JPH0531223Y2 - - Google Patents

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a control device for a variable displacement turbocharger.

[Prior art] A plurality of guide vanes are provided on the inlet side of the turbine of a turbocharger, and when the engine speed is low, the opening degree of the guide vanes is reduced to narrow the area of the nozzle formed by the guide vanes, and when the engine speed is high, the area of the nozzle formed by the guide vanes is narrowed. A control device for a variable displacement turbocharger is known (for example, Japanese Utility Model Application No. 111324/1983), which controls the area of the nozzle to be increased in proportion to the area of the nozzle.

By the way, there is also known a control device that controls the opening degree of the guide vane based on engine rotation and load, and this will be explained with reference to FIG. 3.

The compressor 3 of the turbocharger 2 connected to the engine 1 is provided with a supercharging pressure sensor 4 which is a supercharging pressure detection means and outputs a measured voltage Vb.
The turbine 5 is provided with a vane actuator 6, which is a variable capacity control means for changing the opening degree of a guide vane (not shown).

The control circuit, generally designated by 20, includes a determination circuit 21 that determines the target boost pressure based on signals from the engine rotation sensor 7 and the load sensor 8 and outputs a reference voltage Va; , a comparison circuit 22 that compares the actual measurement value voltage Vb from the boost pressure sensor 4, a control amount calculation circuit 23 that calculates a control amount based on the comparison result of the comparison circuit 22, that is, voltage Va−Vb=Vc, and an amplification It consists of a circuit 24.

To control such a control device, PID control, so-called learning control method, is generally used based on the following calculation formula. That is, Yn=K _P・ΔXn+Σ(K _I・ΔXn) +K _D (Xn−X _o-1 ) () Here, Yn: Controlled amount at that time Xn: Target value at that time X _o-1 : 1 digital ΔXn: Difference between the actual measured value and the target value at that time K _P : Proportionality constant K _I : Integral constant K _D : Differential constant By the way, since the feedback reaction from the boost pressure sensor 4 is slow, the equation ΔXn of () becomes larger, and therefore Yn also becomes larger. This results in so-called over-control, as shown by the chain line in Figure 4.
It is not possible to follow the change in the target boost pressure shown by the solid line, and the actual boost pressure causes hunting, resulting in a decrease in control accuracy.

Therefore, in Utility Application No. 62-40032, the present applicant
We proposed a control system for variable displacement turbochargers that improves control accuracy.

According to the proposal, a target boost pressure determination circuit that determines the target boost pressure from engine rotation and load, a boost pressure detection means that detects the actual boost pressure, and a determination pressure of the target boost pressure determination circuit and the A comparison circuit that determines the difference between the pressure detected by the boost pressure detection means, first and second judgment circuits that determine the supercharging state based on the comparison result of the comparison circuit, and a reference control amount application circuit that provides the obtained reference control amount;
a maximum control amount application circuit, a control amount correction circuit, and a control amount calculation circuit that provide a control amount based on the judgment results of the first and second judgment circuits; the maximum control amount application circuit, the control amount correction circuit, and the control amount calculation circuit; A variable capacity control means is provided for changing the opening degree of the guide vane based on a signal from the circuit.

To explain this with reference to the drawings, in FIG. 5, a control circuit, generally designated by 10, determines a target boost pressure based on detection signals from an engine rotation sensor 7 and an engine load sensor 8, and determines a reference voltage Va. a comparison circuit 12 that determines the difference Vc between the reference voltage Va and the actually measured voltage Vb from the supercharging pressure sensor 4, and a first supercharging circuit that determines the supercharging state based on the voltage difference Vc. A signal is obtained from a three-dimensional map whose parameters are the engine rotation and load, which includes the state judgment circuit 13, the second supercharging state judgment circuit 14, and a reference control amount that allows the target supercharging pressure to be obtained in the standard state of the engine. a reference control amount application circuit that outputs a so-called learning circuit 15; a maximum control amount application circuit 16 that outputs a maximum control amount signal based on the judgment result of the first supercharging state judgment circuit 13; A control amount correction circuit 17 that corrects the reference control amount from the reference control amount application circuit 15 based on the judgment result of the state judgment circuit 14, a control amount calculation circuit 18 that performs calculations on the reference control amount, and these various circuits. 16,
The vane actuator 6 includes an amplifier circuit 19 that amplifies the output signals from the vane actuators 17 and 18 and outputs the amplified signals to the vane actuator 6.

During control, as schematically shown in FIG. 6, when the difference between the target value (indicated by a solid line) and the actual value (indicated by a chain line) is large, the control circuit 10 uses maximum control to bring it as close to the target as possible (region). When the difference reaches a certain value, the amount of control relative to the target is suppressed (region), and when the difference becomes smaller, feedback is applied to achieve fine control without over-controlling (region). Control accuracy can be improved by reducing hunting.

The aspect of this control will be explained in detail. Figure 7 shows the judgment area amounts W3, W1 (minus side) that are on the minus side or plus side with respect to the target value Xn at that time,
A judgment area defined by W4 and W2 (plus side) is provided, and depending on which area the difference ΔXn between the actual measured value and the target value is at that time,
The control amount Yn is taken as shown in FIG.

That is, when the difference ΔXn is in the region ΔXn≧W1, that is, Yn=100% (maximum positive control amount); when the difference ΔXn is in the region ΔXn≦W2, that is, Yn=0% (maximum negative control amount) W3≧ΔXn >W1 In other words, when it is in the region, Yn=Dn+A () where Dn: is the theoretical control amount that gives the target value of Xn,
The correction amount (constant) of A:Dn, which is determined in advance from a map based on the signals from the engine rotation sensor and load sensor, is in the range W2>ΔXn≧W4, that is, Yn=Dn−B () Here B: Correction amount of Dn (constant) W4>ΔXn>W3 In other words, when in the region, Yn=Dn+K _P・ΔX+K _I ΣΔX () This is the theoretical learning control amount corresponding to equation ().

Therefore, A1: Judgment supercharging pressure - equal applied voltage of 200mmHg (quantity W1
) A2: Judgment supercharging pressure - 300mmHg equal applied voltage (quantity W2
) B1: Equal applied voltage of judgment boost pressure - 50 mmHg (corresponds to quantity W3) B2: Equal applied voltage of judged boost pressure - 70 mmHg (corresponds to quantity W4) Set and control as follows. .

That is, the first supercharging state determination circuit 13 determines whether Vc≦A1 or Vc≧A2, and if YES, the maximum control amount application circuit 16 outputs the Theoretical control amount Yn at output
=100% or 0% control. Then, if the above judgment becomes NO, that is, the amount W1 or
When W2 is reached, the second supercharging state judgment circuit 14
Determine whether Vc≦B1 or Vc≧B2.
If YES, in the control amount correction circuit 17,
The theoretical control amount Dn based on the map from the reference control amount application circuit 15 plus correction amount A or correction amount B
The control amount Yn is controlled by subtracting the value Yn. and,
When the above judgment becomes NO, that is, when the amount W3 or the amount W4 is reached, that is, when the area is entered, the controlled amount calculation circuit 18 calculates the controlled amount Yn by the formula ()
This is what is known as learning control. In this state, the actual boost pressure is close to the target value, so changes can be predicted and control can be performed with high accuracy.

[Problem to be solved by the invention] In the device proposed above, the theoretical control amount Dn
is the DUTY ratio expressed from 0 to 100%. Since this Dn is a value for giving the reference voltage Va determined from a certain reference engine, the sliding resistance of the sliding parts of the hardware system of the device, such as the vane shaft in the vane actuator 6 and the turbocharger 2. variations are not taken into account. Therefore Dn
In the region shown in FIG. 7, where the amount of the boost pressure is dominant, the boost pressure is in an equilibrium state, and a problem arises in that the target boost pressure cannot be controlled.

An object of the present invention is to provide a control device for a variable displacement turbocharger that absorbs variations in the hardware of the control device and improves control accuracy.

[Means for Solving the Problems] According to the present invention, there is provided a target boost pressure determination circuit that determines the target boost pressure of the turbine based on signals from an engine rotation sensor and a load sensor, and a circuit that detects the actual boost pressure of the compressor. and a comparison circuit for determining the difference between the determination pressure of the target boost pressure determination circuit and the detection pressure of the boost pressure detection means, In a control device for a variable displacement turbocharger that operates a vane actuator, there is provided a supercharging pressure state determination circuit that determines the supercharging state based on the comparison result of the comparison circuit; When it is determined that the detected pressure is equal to the detected pressure a predetermined time ago, the learning correction amount ( dDn/
The theoretical control amount (Dn ) with a time-series learned correction amount ±Σ(dDn/dt); When the maximum control amount applying circuit that receives a signal and gives a positive or negative maximum control amount when it determines that there is a A signal is input to the reference control amount applying circuit, and the control amount correction circuit corrects the control amount using the signal from the reference control amount application circuit, and the boost pressure state judgment circuit detects that the difference between the judgment pressure and the detected pressure is small. When it is determined that the signal is inputted, a control amount calculation circuit performs learning control using the signal from the reference control amount application circuit, a maximum control amount application circuit, a control amount correction circuit, and a control amount calculation circuit. and an amplifier circuit that amplifies a signal from the turbine and operates the vane actuator of the turbine.

[Function] In the variable displacement turbocharger control device configured as described above, the supercharging state is determined based on the magnitude of the difference between the judgment pressure and the detected pressure, and when the difference is large, the maximum control amount is used to determine the supercharging state. Get closer to the target value,
As the difference becomes smaller, a correction or a value by an arithmetic circuit is added to the reference control amount to quickly bring the actual boost pressure closer to the target boost pressure, which is similar to the proposed device.

In the present invention, further, when the measured voltage Vb and the measured voltage Vb (t) t seconds ago are equal, the theoretical control amount Dn is corrected by a time-series learning correction amount ±Σ(dDn/dt),
Theoretical control amount correction value that absorbs variations in hardware
Obtain DN and complete control promptly. That is, for example, wear and tear on the blades and shaft of the turbocharger can be corrected by this learning correction, and control based on changes over time can be performed correctly. As mentioned above, the theoretical control amount Dn is determined by one engine and the turbocharger, so (a) initially, the instrumental error of the engine turbocharger, that is, the individual variation; (b) Due to the deterioration of the hard parts of the engine and turbocharger during use, the learning correction amount (dDn/dt) is adjusted for each device to satisfy both the time of shipment and subsequent changes over time. It will be adopted. By adopting this learning correction amount (dDn/dt),
Both instantaneous (when t is small) and relatively long time changes can be satisfied.

[Examples] Examples of the present invention will be described below with reference to the drawings.

In FIGS. 1 and 2, parts corresponding to those in FIGS. 3 and 6 are given the same reference numerals, and redundant explanation will be omitted.

In FIG. 1, the comparator circuit 12 has a reference voltage Va
Find the difference Vc between and the measured voltage Vb, and calculate the measured voltage
Vb is compared with the actually measured voltage Vb(t) a predetermined time t seconds ago. In this comparison, it is preferable that the predetermined time t seconds is, for example, 0.5 seconds. This comparison circuit 12 includes a learning correction amount application circuit 2.
0 is connected, and its circuit 20 is connected to the reference control amount application circuit 15. This learning correction amount application circuit 20 receives the actual measured voltage Vb and t from the comparison circuit 12.
The actual measured voltage Vb(t) seconds ago is equal [Vb=vb
(t)] When the signal is output, the learning correction amount dDn/dt
is output to the reference control amount application circuit 15. This learning correction amount dDn/dt is the theoretical control amount
It is preferable to take, for example, 1% per second with respect to Dn.

During control, the control amount Yn is determined as shown in FIG. 2 depending on whether the difference ΔXn between the actual measurement value and the target value is in the range shown in FIG. 7.

That is, when the difference ΔXn is in the region ΔXn≧W1, that is, Yn=100% (maximum positive control amount); when it is in the region ΔXn≦W2, that is, Yn=0% (maximum negative control amount) W3≧ In other words, when in the region ΔXn>W1, Yn=DN+A Here, DN=Dn+Σ(dDn/dt) Dn: Theoretical control amount that gives the target value of Xn A: Correction amount of Dn This DN is the theoretical control of equation (). This is a theoretical control amount correction value obtained by correcting the amount Dn by absorbing variations in the hardware of the device.

In other words, when in the area W2＞ΔXn≥W4, Yn=DN−B where, DN=Dn−Σ(dDn/dt) B: Correction amount of Dn (constant) This DN corresponds to Dn in equation (). This is the theoretical control amount correction value.

W4＞ΔXn＞W3 In other words, when in the area, Yn=DN+K _P・ΔX+K _I ΣΔX () Here, when Vc>0, DN=Dn+Σ(dDn/dt) When Vc<0, DN=Dn−Σ(dDn /dt) This is the theoretical learning control amount correction value corresponding to the above equation ().

Therefore, as mentioned above, A1: Judgment supercharging pressure - equal applied voltage of 200 mmHg (quantity W1
) Etc.A2: Judgment supercharging pressure - 300mmHg equal applied voltage (amount W2
) B1: Equal applied voltage of judgment boost pressure - 50 mmHg (corresponds to quantity W3) B2: Equal applied voltage of judged boost pressure - 70 mmHg (corresponds to quantity W4) Set and control as follows. .

That is, the first supercharging state determining circuit 13 determines whether Vc≦A1 or Vc≧A2, and if YES, the maximum control amount applying circuit 16 outputs the Theoretical control amount Yn at output
=100% or 0% control. Then, if the above judgment becomes NO, that is, the amount W1 or
When W2 is reached, the second supercharging state judgment circuit 14
Determine whether Vc≦B1 or Vc≧B2.
If YES, in the control amount correction circuit 17,
The control amount Yn is calculated by adding the correction amount A or minus the correction amount B to the theoretical control amount correction value DN, which is obtained by correcting the theoretical control amount based on the map from the reference control amount application circuit 15 with the learning correction amount of the learning correction amount application circuit. Take control. And if the above judgment becomes NO,
In other words, when the amount W3 or the amount W4 is reached, that is, when it enters the region, the controlled amount calculation circuit 18 calculates the controlled amount.
Yn is calculated by formula () and so-called learning control is performed.

[Effects of the invention] Since the present invention is configured as described above, it produces the effects described below.

In other words, when the difference between the measured pressure and the target pressure is large, the maximum control is used to bring it as close to the target pressure as possible, and as the difference becomes smaller, the reference control amount is corrected or a value is added by the calculation circuit to quickly adjust the target pressure. The actual measured pressure can be brought close to .

In addition, it is possible to absorb variations in the hardware of the device, correct the theoretical control amount, and further improve control accuracy.

Since the theoretical control amount is learned and corrected in this way, it is possible to accurately control changes over time due to, for example, wear of turbocharger blades and shafts.

[Brief explanation of the drawing]

FIG. 1 is an overall configuration diagram showing an embodiment of the present invention;
Fig. 2 is a control characteristic diagram explaining the effects of the present invention, Fig. 3 is an overall configuration diagram showing a conventional device, Fig. 4 is a control characteristic diagram thereof, and Fig. 5 is a diagram showing the device previously proposed by the present applicant. 6 is a control characteristic diagram illustrating the outline of the effect, FIG. 7 is a diagram illustrating the determination area, and FIG. 8 is a control characteristic diagram illustrating the effect in detail. 4...Supercharging pressure sensor, 6...Vane actuator, 10...Control circuit, 11...Judgment circuit, 1
2... Comparison circuit, 13... First supercharging state determining circuit, 14... Second supercharging state determining circuit, 15...
Reference control amount application circuit, 16... Maximum control amount application circuit, 18... Controlled amount calculation circuit, 20... Learning correction amount application circuit.

Claims (1)

[Scope of utility model registration request] a target boost pressure determination circuit that determines the target boost pressure of the turbine based on signals from an engine rotation sensor and a load sensor; a boost pressure detection means that detects the actual boost pressure of the compressor; and the target boost pressure determination circuit. A control device for a variable displacement turbocharger, comprising: a comparison circuit for determining a difference between a judgment pressure of and a detection pressure of the supercharging pressure detection means, and operates a vane actuator of a turbine based on a comparison result of the comparison circuit, a supercharging pressure state determination circuit that determines a supercharging state based on a comparison result of the comparison circuit; and a supercharging pressure state determination circuit that determines that the latest detected pressure of supercharging pressure detection is equal to the detected pressure a predetermined time ago by the comparison circuit. a learning correction amount application circuit that calculates a learning correction amount (dDn/dt) that is a change amount per predetermined time in the theoretical control amount (Dn) determined by the signals from the engine rotation sensor and the load sensor; The theoretical control amount (Dn) is calculated as a time-series learning correction amount ±Σ(dDn /dt), and furthermore, when the boost pressure state judgment circuit judges that the difference between the judgment pressure and the detected pressure is large, a signal is inputted to output a positive or negative signal. A signal is input when the maximum control amount application circuit that provides the maximum control amount and the supercharging pressure state judgment circuit determine that the difference between the judgment pressure and the detected pressure is medium, and the reference control amount application is performed. A signal is input when the control amount correction circuit corrects the control amount using a signal from the circuit and the supercharging pressure state judgment circuit judges that the difference between the judgment pressure and the detected pressure is small; a control amount calculation circuit that performs learning control using the signal from the reference control amount application circuit;
Control of a variable displacement turbocharger characterized by comprising an amplifier circuit that amplifies signals from the maximum control amount application circuit, the control amount correction circuit, and the control amount calculation circuit to operate the vane actuator of the turbine. Device.