JP2016075174A - Turbine efficiency learning processing method and boost pressure control device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To enable a more appropriate learning value taking to be acquired as compared with that of the prior art by executing learning processing in response to an accuracy of model control.SOLUTION: An electronic control unit operates as follows. A difference between a model opening that is a high pressure wastegate opening calculated by calculation based on a model control and a duty of PWM signal that is a drive signal for a high pressure wastegate corresponding to an actual opening of the high pressure wastegate is multiplied by a model learning factor and converted into a correction value of a turbine efficiency, then, the corrected values of the turbine efficiency are integrated, a result of the integration is applied to a boost pressure control based on the model control as a learning value for the corrected value of the turbine efficiency, and in turn, the model learning factor is read out of a model learning factor map constituted in such a way that a pressure ratio and a flow rate of the high pressure compressor can be read as input parameters and in turn the learning processing is stopped in response to the pressure ratio and the flow rate of the high pressure compressor.SELECTED DRAWING: Figure 3

Description

  The present invention relates to a turbine efficiency learning processing method and a supercharging pressure control device in a supercharging pressure control device that performs operation control of a supercharger that supercharges intake air of an internal combustion engine, and in particular, along with improving the reliability of learning processing. , Related to improvements in drivability.

Since the supercharger is an effective means for emission control by forcibly sending air into the internal combustion engine, various methods, devices, etc. have been proposed and put to practical use as a method and device for controlling the supercharging pressure. As is well known in the art.
As one of such supercharging pressure devices, two high pressure superchargers and low pressure superchargers are provided from the viewpoint of enabling more appropriate supercharging pressure control according to the running state of the vehicle. A so-called two-stage supercharging system has been proposed and put into practical use (see, for example, Patent Document 1).

  As a method of supercharging pressure control in such a supercharging system, a so-called supercharger system including an intake pipe and an exhaust pipe connecting a supercharger and a supercharger to an internal combustion engine is modeled based on thermodynamics and the like. The actual superacquired pressure data is input to the model to obtain the control target value (model value) of the turbocharger system. So-called model control for controlling the above has been proposed and put into practical use (see, for example, Patent Document 2).

  In such a supercharging pressure control device using model control, in order to enable appropriate supercharging pressure control to be executed even if there is a variation in hardware among individual devices or a variation in characteristics of various sensor elements. For example, a learning process for performing turbine efficiency is executed, and the learned value is used for supercharging pressure control based on model control.

JP 2007-138845 A (page 5-9, FIGS. 1 to 13) JP 2009-168007 A (page 4-18, FIG. 1 to FIG. 18)

  By the way, in the two-stage supercharging system as described above, both the high-pressure and low-pressure turbochargers are operated (two-stage operation) when the flow rate of the intake air is low, and the low pressure is high when the flow rate is high. Only the turbocharger for operation is operated (one-stage operation), but in order to smooth the torque fluctuation at the time of switching, the second stage after the high-pressure supercharger stops working A technique of switching from operation to one-stage operation may be employed. In such a case, the high-pressure compressor enters the choke region that is the supercharging limit, that is, the state of high flow rate and the pressure ratio before and after the high-pressure compressor = 1, but the accuracy of the model control The supercharging pressure control based on the model control may be effective even in such a so-called boundary region where the pressure decreases.

  On the other hand, the turbine efficiency learning process described above is executed even if the operation state as described above is in the boundary region if the model control is effectively applied to the supercharging pressure control. In addition, a value that should not be a learning value is sometimes acquired as a learning value, and as a result, there is a problem in that overshooting or undershooting of the supercharging pressure is caused.

  The present invention has been made in view of the above circumstances, and is a turbine in a supercharging pressure control device that enables acquisition of a more appropriate learning value as compared to the conventional one by executing learning processing according to the accuracy of model control. An efficiency learning processing method and a supercharging pressure control device are provided.

In order to achieve the object of the present invention, a turbine efficiency learning processing method according to the present invention includes:
The electronic control unit executes the supercharging pressure of the intake air in a two-stage supercharging system configured to pressurize the intake air of the internal combustion engine using a high-pressure turbocharger and a low-pressure turbocharger and send the pressurized air to the internal combustion engine. A turbine efficiency learning processing method in a supercharging pressure control device configured to be controllable based on model control to be performed,
An opening deviation that is a difference between a model opening that is a calculated opening of the high-pressure wastegate constituting the high-pressure turbocharger and an actual opening of the high-pressure wastegate that is calculated by calculation in the model control Is multiplied by a model learning factor to convert it to a turbine efficiency correction value, and then integration is performed on the turbine efficiency correction value, and the integration result is used as a learning value for the turbine efficiency correction value for the model control. For supercharging pressure control based on
The model learning factor is read from a model learning factor map configured to be readable with the pressure ratio and flow rate of the high-pressure compressor constituting the high-pressure turbocharger as input parameters, while the pressure ratio of the high-pressure compressor and The learning process can be stopped according to the flow rate.
In order to achieve the above object of the present invention, a supercharging pressure control device according to the present invention includes:
The electronic control unit executes the supercharging pressure of the intake air in a two-stage supercharging system configured to pressurize the intake air of the internal combustion engine using a high-pressure turbocharger and a low-pressure turbocharger and send the pressurized air to the internal combustion engine. A supercharging pressure control device configured to be controllable based on model control to be performed,
The electronic control unit is
An opening deviation that is a difference between a model opening that is a calculated opening of the high-pressure wastegate constituting the high-pressure turbocharger and an actual opening of the high-pressure wastegate that is calculated by calculation in the model control Is multiplied by a model learning factor to convert it to a turbine efficiency correction value, and then integration is performed on the turbine efficiency correction value, and the integration result is used as a learning value for the turbine efficiency correction value for the model control. The model learning factor is read from a model learning factor map configured to be readable as an input parameter of a pressure ratio and a flow rate of a high-pressure compressor that constitutes the high-pressure turbocharger. Turbo efficiency learning process that can stop the learning process according to the pressure ratio and flow rate of the high-pressure compressor can be executed Is made is configured.

  According to the present invention, the learning process is stopped in the area where the accuracy of the model control is lowered. Therefore, unlike the conventional technique, the acquisition of an inappropriate learning value or an incorrect learning value is surely prevented. Since the turbine efficiency is properly corrected, the occurrence of overshooting and undershooting of the boost pressure due to improper learning values or incorrect learning values is suppressed and prevented, and more stable. Thus, the supercharging pressure control based on the highly reliable model control can be realized, and as a result, it is possible to contribute to the improvement of drivability.

It is a block diagram which shows the structural example of the supercharging pressure control apparatus in embodiment of this invention. It is a block diagram which shows the function which an electronic control unit performs when the turbine efficiency learning control process in embodiment of this invention is performed in the electronic control unit which comprises the supercharging pressure control apparatus shown by FIG. 3 is a subroutine flowchart showing a procedure of a turbine efficiency learning process in the embodiment of the present invention executed by an electronic control unit constituting the supercharging pressure control device shown in FIG. 1.

Hereinafter, embodiments of the present invention will be described with reference to FIGS. 1 to 3.
The members and arrangements described below do not limit the present invention and can be variously modified within the scope of the gist of the present invention.
First, the configuration of the supercharging pressure control device in the embodiment of the present invention will be described with reference to FIG.
The supercharging pressure control device according to the embodiment of the present invention is configured by a so-called two-stage supercharging system having two superchargers. A turbocharger (hereinafter referred to as “turbocharger”) is used.

  That is, first, in an engine 1 using a diesel engine, an intake pipe 2 for taking in air necessary for combustion exhausts exhaust gas from the engine 1 to an intake manifold (not shown) of the engine 1. For this purpose, exhaust pipes 3 are connected to exhaust manifolds (not shown).

A high-pressure turbocharger 4 and a low-pressure turbocharger 5 are provided at appropriate positions in the middle of the piping of the intake pipe 2 and the exhaust pipe 3 with appropriate intervals from the engine 1 side.
Both the high-pressure turbocharger 4 and the low-pressure turbocharger 5 have the same basic configuration as that of a conventional turbocharger.
That is, the high-pressure turbocharger 4 is configured with the high-pressure turbine 6 and the high-pressure compressor 7 as main components, and the low-pressure turbocharger 5 is configured with the low-pressure turbine 8 and the low-pressure compressor 9 as main components. It has been made.

  In the embodiment of the present invention, the high-pressure turbine 6 uses a variable nozzle turbo (not shown). Such a variable nozzle turbo has a well-known structure in which a variable nozzle mechanism is provided in a turbine.

The high-pressure turbine 6 and the low-pressure turbine 8 are spaced at appropriate positions in the middle of the exhaust pipe 3 from the engine 1 side to the downstream side in the order of the high-pressure turbine 6 and the low-pressure turbine 8. Is provided.
The high-pressure compressor 7 and the low-pressure compressor 9 are arranged at appropriate positions in the intake pipe 2 in the order of the high-pressure compressor 7 and the low-pressure compressor 9 from the engine 1 side to the upstream side. They are provided at an appropriate interval.

A high-pressure wastegate 11 is provided before and after the high-pressure turbine 6, and a low-pressure wastegate 12 is provided before and after the low-pressure turbine 8. By adjusting the variable nozzle 6, the flow rate of exhaust gas flowing into the high-pressure turbine 6 and the low-pressure turbine 8, in other words, the supercharging pressure can be controlled.
In addition, a bypass valve 13 is provided before and after the high-pressure compressor 7 so that the high-pressure compressor 7 can be bypassed as necessary.

The intake pipe 2 and the exhaust pipe 3 are communicated with each other by a communication pipe 14 provided at an appropriate location between the high-pressure turbocharger 4 and the engine 1.
An exhaust gas recirculation valve 15 is provided at an appropriate position in the middle of the communication pipe 14, and the return amount of the exhaust gas from the exhaust pipe 3 to the intake pipe 2 can be varied by adjusting the opening thereof. ing.

  Further, in the intake pipe 2, the intercooler 16 that cools the intake air is adjusted from the high-pressure compressor 7 side to an appropriate position between the high-pressure compressor 7 and the communication pipe 14, and the amount of intake air is adjusted. An intake throttle valve 17 is provided in order at an appropriate interval. Then, at an appropriate position of the intake pipe 2 between the intake throttle valve 17 and the communication pipe 14, from the intake throttle valve 17 side, a supercharging pressure sensor 18 for detecting the supercharging pressure, the temperature of the intake air of the engine 1 An intake air temperature sensor 19 for detecting the above is provided in order at an appropriate interval.

The variable nozzle turbo (not shown), the high-pressure wastegate 11 and the low-pressure wastegate 12, the bypass valve 13, the exhaust gas recirculation valve 15, and the intake throttle valve 17, which constitute the high-pressure turbine 6 described above, The opening degree is controlled by the electronic control unit 101.
The electronic control unit 101 includes, for example, a microcomputer having a known and well-known configuration, a storage element (not shown) such as a RAM and a ROM, and an input / output interface circuit (not shown). It is configured as a main component.

The electronic control unit 101 executes control processing by various software necessary for operation control of the engine 1, and as one of the control processing, turbine efficiency learning in an embodiment of the present invention to be described later is performed. Processing is to be executed.
In the electronic control unit 101, in addition to the detection signals of the supercharging pressure sensor 18 and the intake air temperature sensor 19, the atmospheric pressure, the engine speed, the accelerator opening, the engine coolant temperature, etc. detected by a sensor (not shown) or the like. It is input and used for engine operation control processing, turbine efficiency learning processing in an embodiment of the present invention to be described later, and the like.

Next, the turbine efficiency learning process in the embodiment of the present invention executed by the electronic control unit 101 will be described with reference to FIGS. 2 and 3.
First, the precondition of the supercharging pressure control apparatus in which the turbine efficiency learning process in the embodiment of the present invention is executed will be described.
The supercharging pressure control apparatus in the embodiment of the present invention is premised on that the supercharging pressure control is configured to be executed based on so-called model control.

  In other words, the supercharging pressure control based on the model control is performed by modeling the supercharging system including the high-pressure turbocharger 4 and the low-pressure turbocharger 5 and the intake pipe 2 and the exhaust pipe 3 based on the thermodynamics. The obtained supercharging pressure and other data is input to the model to obtain the model value that is the control target value of the supercharging system, and the operation control is performed so that the operating state of the turbocharger reaches the state of the model value. It is based on what to do.

  Here, the physical quantity used as the control target value (model value) is, for example, the supercharging pressure or the rotation speed of the turbocharger. What is used as the model value is a model based on thermodynamics or the like. Depending on how it is set, it does not need to be limited to a specific one. The supercharging system is modeled based on thermodynamics, etc., and the supercharging pressure is controlled based on the model. If there is, it is good.

In the embodiment of the present invention, the electronic control unit 101 calculates the target supercharging pressure and the target opening of each of the high pressure wastegate 11 and the low pressure wastegate 12 by calculation based on model control. Operation control is performed so that the pressure and the target opening are obtained.
That is, the supercharging pressure control in the supercharging pressure control device according to the embodiment of the present invention is such that when the engine 1 is in a low rotation state, the two turbochargers 4 and 5 are driven by model control, while the engine 1 When 1 is in a high rotation state, only the low-pressure side low-pressure turbocharger 5 is basically driven by model control.

Furthermore, in the embodiment of the present invention, when the engine 1 is in a high rotation state and only the low-pressure side low-pressure turbocharger 5 is driven and controlled by model control, the high-pressure turbocharger 4 does not work completely. Thus, it is assumed that only the low-pressure turbocharger 5 is switched to the driven state.
Hereinafter, the state in which the two turbochargers 4 and 5 are driven as required is referred to as “two-stage operation” or “two-stage operation”, and the state in which only the low-pressure turbocharger 5 is driven is “ It will be referred to as “one stage operation” or “one stage operation”.

Next, the turbine efficiency learning process executed by the electronic control unit 101 under such a premise will be described with reference to the block diagram shown in FIG. 2 and the subroutine flowchart shown in FIG.
First, the turbine efficiency learning process in the embodiment of the present invention will be generally described with reference to the block diagram shown in FIG.
In learning the turbine efficiency, first, the model opening of the high-pressure wastegate 11 (indicated as “HP WG” in FIG. 2) and the duty of the PWM signal that drives the high-pressure wastegate 11 (hereinafter, for convenience of explanation). (Hereinafter referred to as “opening deviation” for convenience of explanation).
The turbine efficiency is obtained by dividing the work performed by the turbine by the exhaust energy input to the turbine. In the embodiment of the present invention, the calculated value in the model control is obtained as follows. The value is calculated as the total turbine efficiency of the high-pressure turbine 6 and the low-pressure turbine 8.

Here, the model opening of the high-pressure wastegate 11 is a theoretical opening of the high-pressure wastegate 11 calculated by a calculation process in model control.
In the embodiment of the present invention, the high-pressure wastegate 11 is driven by a PWM signal as in the prior art, and the opening degree is determined by the duty. Therefore, the high-pressure wastegate duty is the high-pressure wastegate 11. It corresponds to the actual opening.

  Therefore, the opening deviation that is the difference between the model opening of the high-pressure wastegate 11 and the high-pressure wastegate duty is the difference between the model opening of the high-pressure wastegate 11 and the actual opening of the high-pressure wastegate 11. Corresponding to the difference, it represents an error in the opening of the high-pressure wastegate 11 and is expressed in percentage. In the embodiment of the present invention, in view of the opening degree error in model control, the opening degree deviation is referred to as “model opening degree error” (see FIG. 2).

  In order to replace this model opening error with the dimension of turbine efficiency, conventionally, a predetermined coefficient is multiplied, and an integral value of the multiplication result is used as a learned value (turbine efficiency model learned value) related to turbine efficiency in model control. It was supposed to be obtained. Since the turbine efficiency model learning value is based on the opening degree deviation as described above, it is not the turbine efficiency itself, but the difference between the turbine efficiency obtained by calculation in model control and the actual turbine efficiency. Therefore, it is used for correction of turbine efficiency.

On the other hand, in the embodiment of the present invention, the pressure ratio of the high-pressure compressor (indicated as “HP compressor” in FIG. 2) 7 and the high-pressure compressor 7 against the model opening error described above. The product is multiplied by a model learning factor (model learning factor) determined according to the flow rate, and the multiplication result is integrated as in the conventional case.
In view of the fact that the supercharging pressure control by the model control is not always appropriate in all operating states, the model learning factor is not limited to the turbine efficiency model even in the region where the accuracy of the supercharging pressure control by the model control decreases. In order to make the learning value appropriate, a value corresponding to the pressure ratio of the high-pressure compressor 7 and the flow rate of the high-pressure compressor 7 is determined by the model learning factor map 51 (see FIG. 2). .

The model learning factor map is a model learning factor for various combinations of the pressure ratio of the high-pressure compressor 7 and the flow rate of the high-pressure compressor 7, and the pressure ratio of the high-pressure compressor 7 and the flow rate of the high-pressure compressor 7 are input. The model learning factor for the combination of various pressure ratios of the high-pressure compressor 7 and the flow rate of the high-pressure compressor 7 is determined based on test results and simulation results. It is.
Such a model learning factor map is stored in advance in an appropriate storage area of the electronic control unit 101.

Here, the pressure ratio of the high-pressure compressor 7 and the flow rate of the high-pressure compressor 7 are both model calculation values calculated by predetermined arithmetic processing in model control.
In particular, in the embodiment of the present invention, in a specific case as described later, the model learning factor is set to “0”, thereby substantially stopping the learning process, Unlike the prior art, inadequate acquisition of the turbine efficiency model learning value is avoided in an area where the accuracy of model control is reduced.

Next, the procedure of the turbine efficiency learning process executed by the electronic control unit 101 in the embodiment of the present invention will be described with reference to the subroutine flowchart shown in FIG.
When the processing by the electronic control unit 101 is started, the pressure ratio of the high-pressure compressor 7 and the flow rate of the high-pressure compressor 7 at this time are acquired (see step S102 in FIG. 3).
Here, the pressure ratio of the high-pressure compressor 7 and the flow rate of the high-pressure compressor 7 are calculated values calculated and calculated in supercharging pressure control processing based on model control executed in a main routine (not shown). In this step S102, there is no need to calculate again, and the value already calculated in the main routine can be used.

Next, the model learning factor is read (see step S106 in FIG. 3).
That is, the model in which the pressure learning ratio of the high-pressure compressor 7 and the model learning factor for the flow rate of the high-pressure compressor 7 acquired in step S102 are stored in advance in an appropriate storage area of the electronic control unit 101. Read from the learning factor map 51.
Next, it is determined whether or not the read model learning factor is “0” (see step S106 in FIG. 3).

In the embodiment of the present invention, in order to avoid acquisition of an inappropriate turbine efficiency model learning value in a region where the accuracy of model control is reduced, the high pressure compressor 7 used for reading the model learning factor is used. The model learning factor of the model learning factor map 51 is “0” when the pressure ratio is equal to or lower than the predetermined pressure ratio and the flow rate of the high-pressure compressor 7 used for reading the model learning factor is equal to or higher than the predetermined flow rate. It is set to "".
Here, the predetermined pressure ratio and the predetermined flow rate are appropriate values for avoiding acquisition of an inappropriate turbine efficiency model learning value in consideration of operation characteristics of individual model control, vehicle specifications, and the like. It is preferable to set based on test results and simulation results.

  Therefore, when it is determined in step S106 that the model learning factor is “0” (in the case of YES), the product of the model opening error and the model learning factor described in FIG. Because it becomes “0”, the increment of the most recent integral value for the product of the model opening error and the model learning factor is “0”, and the integral value is maintained, and a new turbine efficiency model learning value is acquired. In other words, the model learning process is substantially stopped (see FIG. 2).

On the other hand, when it is determined in step S106 that the model learning factor is not “0” (in the case of NO), turbine efficiency model learning is executed as usual, and a turbine efficiency model learning value is acquired. (See FIG. 2).
As described above, in the embodiment of the present invention, unlike the conventional case, since acquisition of an inappropriate turbine efficiency model learning value is avoided, the supercharging pressure based on model control using the inappropriate turbine efficiency model learning value is avoided. As a result of the control, the occurrence of overshoot and undershoot of the boost pressure is reliably suppressed and prevented, and the boost pressure control based on the model control that is more stable than before is realized. .

  The present invention can be applied to a supercharging pressure control device in which it is desired to acquire a more appropriate learning value in a learning process in supercharging pressure control based on model control.

DESCRIPTION OF SYMBOLS 1 ... Engine 4 ... High pressure turbocharger 5 ... Low pressure turbocharger 11 ... High pressure wastegate 12 ... Low pressure wastegate 18 ... Supercharging pressure sensor 101 ... Electronic control unit

Claims (6)

The electronic control unit executes the supercharging pressure of the intake air in a two-stage supercharging system configured to pressurize the intake air of the internal combustion engine using a high-pressure turbocharger and a low-pressure turbocharger and send the pressurized air to the internal combustion engine. A turbine efficiency learning processing method in a supercharging pressure control device configured to be controllable based on model control to be performed,
An opening deviation that is a difference between a model opening that is a calculated opening of the high-pressure wastegate constituting the high-pressure turbocharger and an actual opening of the high-pressure wastegate that is calculated by calculation in the model control Is multiplied by a model learning factor to convert it to a turbine efficiency correction value, and then integration is performed on the turbine efficiency correction value, and the integration result is used as a learning value for the turbine efficiency correction value for the model control. For supercharging pressure control based on
The model learning factor is read from a model learning factor map configured to be readable with the pressure ratio and flow rate of the high-pressure compressor constituting the high-pressure turbocharger as input parameters, while the pressure ratio of the high-pressure compressor and A turbine efficiency learning processing method, wherein learning processing can be stopped according to a flow rate.   The model learning factor map is obtained by setting the model learning factor to zero in a region where the pressure ratio of the high pressure compressor is equal to or lower than a predetermined pressure ratio and the flow rate of the high pressure compressor is equal to or higher than a predetermined flow rate. The turbo efficiency learning processing method according to claim 1, wherein the learning process is configured to be able to be stopped.   3. The turbo efficiency learning processing method according to claim 2, wherein a duty of a PWM signal that is a drive signal of the high-pressure wastegate is used as an actual opening of the high-pressure wastegate. The electronic control unit executes the supercharging pressure of the intake air in a two-stage supercharging system configured to pressurize the intake air of the internal combustion engine using a high-pressure turbocharger and a low-pressure turbocharger and send the pressurized air to the internal combustion engine. A supercharging pressure control device configured to be controllable based on model control to be performed,
The electronic control unit is
An opening deviation that is a difference between a model opening that is a calculated opening of the high-pressure wastegate constituting the high-pressure turbocharger and an actual opening of the high-pressure wastegate that is calculated by calculation in the model control Is multiplied by a model learning factor to convert it to a turbine efficiency correction value, and then integration is performed on the turbine efficiency correction value, and the integration result is used as a learning value for the turbine efficiency correction value for the model control. The model learning factor is read from a model learning factor map configured to be readable as an input parameter of a pressure ratio and a flow rate of a high-pressure compressor that constitutes the high-pressure turbocharger. Turbo efficiency learning process that can stop the learning process according to the pressure ratio and flow rate of the high-pressure compressor can be executed Boost pressure control apparatus characterized by comprising been configured.   The model learning factor map is obtained by setting the model learning factor to zero in a region where the pressure ratio of the high pressure compressor is equal to or lower than a predetermined pressure ratio and the flow rate of the high pressure compressor is equal to or higher than a predetermined flow rate. The supercharging pressure control device according to claim 4, wherein the learning processing is configured to be stopped.   6. The turbo efficiency learning processing method according to claim 5, wherein the electronic control unit uses a duty of a PWM signal that is a drive signal of the high-pressure wastegate as an actual opening of the high-pressure wastegate.

Images