EP1745204A1

EP1745204A1 - Method for pressure regulation of an accumulator injection system

Info

Publication number: EP1745204A1
Application number: EP05740728A
Authority: EP
Inventors: Armin DÖLKER
Original assignee: MTU Friedrichshafen GmbH
Current assignee: Rolls Royce Solutions GmbH
Priority date: 2004-05-12
Filing date: 2005-05-10
Publication date: 2007-01-24
Also published as: US7270115B2; DE102004023365A1; DE102004023365B4; US20070056561A1; WO2005111402A1

Abstract

The invention relates to a pressure regulation method for an internal combustion engine (1) having an accumulator injection system. According to said method, a first actual rail pressure is determined via a first filter from the measured rail pressure (pCR), a first deviation is calculated by comparing the desired/actual rail pressure, and a volume flow is determined as the regulating variable via a high-pressure regulator from the first deviation. According to the invention, a second actual rail pressure is determined from the measured rail pressure (pCR) via a second filter and said second actual pressure is made the reference value for calculating the regulator portions of the high-pressure regulator.

Description

Process for regulating the pressure of a storage injection system

The invention relates to a method for pressure control of a storage injection system of an internal combustion engine according to the preamble of claim 1.

In a storage injection system, a high-pressure pump delivers the fuel from a fuel tank to a rail. The inlet cross-section to the high-pressure pump is determined via a variable suction throttle. Connected to the rail are injectors via which the fuel is injected into the combustion chambers of the internal combustion engine. Since the quality of the combustion depends crucially on the pressure level in the rail, this is regulated. The high-pressure control circuit includes a high-pressure controller, the suction throttle with high-pressure pump and the rail as a control system, as well as a filter in the feedback branch. Typically, the high pressure controller is designed as a PID controller or PIDTl controller, i. H. this includes at least a proportional component (P component), an integral component (I component) and a differential component (D component). In this high-pressure control loop, the pressure level in the rail corresponds to the controlled variable. The measured pressure values of the rail are converted into an actual rail pressure via the filter and compared with a target rail pressure. The resulting control deviation is converted into a control signal for the suction throttle via the high-pressure controller. The control signal corresponds to z. B. a volume flow with the unit liter / minute. The high pressure control circuit described above is known from the unpublished German patent application with the official file number DE 103 30 466.5.

To protect against excessive pressure levels, a passive pressure relief valve is installed on the rail. With one too high pressure level opens the pressure relief valve, whereby the fuel is diverted from the rail into the fuel tank.

In practice, the following problem can occur: The engine speed increases immediately when a load is shed. An increasing engine speed results in a speed control deviation that increases in amount at a constant target speed. A speed controller responds to this by reducing the injection quantity as a manipulated variable. A lower injection quantity in turn means that less fuel is drawn from the rail and therefore the pressure level in the rail increases rapidly. To make matters worse, the delivery rate of the high-pressure pump is speed-dependent. An increasing engine speed means a higher delivery rate and thus causes an additional pressure increase in the rail. Since the high-pressure control has a long response time, the rail pressure can rise so far that the pressure relief valve opens, e.g. B. at 1950 bar. Then the rail pressure drops z. B. to a value of 800 bar. At this pressure level, a state of equilibrium is established between the fuel delivered and the fuel derived. This means that despite the pressure relief valve being opened, the rail pressure does not drop any further. The pressure limiting valve only closes again when the engine speed is reduced. The unexpected opening of the pressure relief valve during load shedding is therefore problematic.

The object of the invention is to improve the safety of the pressure control.

The object is solved by the features of claim 1. The configurations are shown in the subclaims. The invention provides that a second actual rail pressure is determined from the measured rail pressure via a second filter and the second actual rail pressure is set as decisive for the calculation of the regulator components of the high pressure regulator. Controller components are to be understood as the P component, I component, D component and DTl component. The second filter has a smaller time constant and a smaller phase delay than the first filter in the feedback branch. The central idea of the invention is therefore to increase the dynamics of the high-pressure control loop by introducing a second “fast” filter.

To shorten the reaction time, the invention provides that a first proportional coefficient for determining the P component and a first retention time for determining the D component of the high-pressure regulator are each calculated using a characteristic curve as a function of the second actual rail pressure. For this purpose, the characteristic curves have a static and a dynamic range. In the dynamic range, an increasing second actual rail pressure is assigned a likewise increasing proportional coefficient or an increasing lead time via the characteristic curves.

In one embodiment of the invention, it is provided that a second control deviation is calculated from the target rail pressure and the second actual rail pressure, and the P component, D component and DTI component of the high-pressure regulator are determined as a function of the second control deviation , The proportional coefficient for the P component, the retention time for the D component and the DTI component are calculated using the corresponding characteristic curves.

An unexpected opening of the pressure relief valve is reliably prevented by the invention by the dynamics of the high pressure control loop is increased. The consequence is a continuous operation of the internal combustion engine even with a significant change in load on the output, z. B. with a generator load cut-off or when a ship propulsion system emerges.

The preferred exemplary embodiments of the invention are illustrated in the drawings. Show it:

1 shows a system diagram;

2 shows a high-pressure control circuit, first embodiment;

3 shows a first characteristic curve;

Fig. 4 shows a second characteristic;

5 shows a high-pressure control circuit, second embodiment;

Fig. 6 is a block diagram, proportional coefficient;

7 shows a block diagram, lead time;

Fig. 8 is a block diagram, volume flow;

Fig. 9 is a block diagram, DTl portion.

Figure 1 shows a system diagram. In the internal combustion engine 1 shown, the fuel is injected via a storage injection system (common rail). This comprises the following components: a high-pressure pump 3 with a suction throttle for delivering the fuel from a fuel tank 2, a rail 6 for storing the fuel and injectors 7 for injecting the fuel from the rail 6 into the combustion chambers of the internal combustion engine 1.

The operating mode of the internal combustion engine 1 is regulated by an electronic control unit (ADEC) 4. The electronic control unit 4 contains the usual components of a microcomputer system, for example a microprocessor, I / O modules, buffers and memory modules (EEPROM, RAM). The memory modules are used for the operation of the combustion engine 1 relevant operating data applied in maps / characteristic curves. The electronic control unit 4 uses this to calculate the output variables from the input variables. The following input variables are shown by way of example in FIG. 1: a rail pressure pCR, which is measured by means of a rail pressure sensor 5, a speed signal nMOT of the internal combustion engine 1, a signal FP for power specification by the operator and an input variable E. Below the input variable For example, the charge air pressure of a turbocharger and the temperatures of the coolants / lubricants and the fuel are subsumed.

In FIG. 1, a signal ADV for controlling the suction throttle and an output variable A are shown as output variables of the electronic control unit 4. The output variable A represents the other control signals for controlling and regulating the internal combustion engine 1, for example the start of injection SB and the duration of injection SD. In practice, the ADV signal is designed as a pulse-width-modulated signal (PWM).

Such a memory injection system is at a maximum stationary rail pressure of z. B. operated 1800 bar. To protect the system from an impermissibly high pressure level in the rail 6, a passive pressure relief valve 8 is provided. This opens at a pressure level of z. B. 1950 bar. In the open state, the fuel is discharged from the rail 6 into the fuel tank 2 via the pressure relief valve 8. As a result, the pressure level in the rail 6 drops to a value of z. B. 800 bar.

FIG. 2 shows a high-pressure control circuit for regulating the rail pressure pCR in a first embodiment. The input variable corresponds to the setpoint of the rail pressure pCR (SL). The output quantity corresponds to the raw value of the Rail pressure pCR. A first actual rail pressure pCRl (IST) is determined from the raw values of the rail pressure pCR via a first filter 13. This is compared with the setpoint pCR (SL) at a summation point A, which results in a first control deviation dRl. A manipulated variable is calculated from the first control deviation dRI by means of a high-pressure controller 9. The manipulated variable corresponds to a volume flow qV. The physical unit of the volume flow can, for. B. liters / minute. It can optionally be provided that the calculated target consumption is added to the volume flow qV. The volume flow qV corresponds to the input variable for a limitation 10. The limitation 10 can be speed-dependent, input variable nMOT. The output variable qV (SL) of the limitation 10 is then converted into a PWM signal in a function block 11. Fluctuations in the operating voltage and the fuel admission pressure are taken into account in the conversion, input variable E. The PWM signal ADV is then applied to the solenoid of the suction throttle. As a result, the path of the magnetic core is changed, whereby the flow rate of the high-pressure pump 3 is freely influenced. The high-pressure pump 3 with suction throttle and the rail 6 correspond to the controlled system 12. A volume flow qV (VER) is discharged from the rail 6 via the injectors 7. This closes the control loop.

According to the invention, it is provided that a second actual rail pressure pCR2 (IST) is calculated from the raw values of the rail pressure pCR via a second filter 14. The second filter 14 has a smaller time constant and thus a smaller phase delay than the first filter 13. This means that the second actual rail pressure pCR2 (IST) is less time-delayed than the first actual rail pressure pCRl (IST). The calculation of the controller components of the high-pressure controller 9 is significantly influenced by the second actual rail pressure pCR2 (IST). For this purpose, in the embodiment according to FIG. 2, the second actual rail pressure pCR2 (IST) is fed directly to the high-pressure regulator 9.

A first characteristic curve 15 is shown in FIG. A first proportional coefficient kpl is determined via the first characteristic curve 15 for determining a P component of the high-pressure regulator 9. The second actual rail pressure pCR2 (IST) is plotted in bar on the abscissa. The first proportional coefficient kpl is plotted on the ordinate as the output variable. The first characteristic curve 15 comprises a stationary area STAT and a dynamic area DYN. The stationary area ends at a pressure value of 1800 bar. This corresponds to the maximum stationary rail pressure at full load. The dynamic range DYN starts at a pressure value of 1820 bar. To increase security, a tolerance band TB is provided between the stationary and the dynamic range, e.g. B. 20 bar. The first characteristic curve 15 is composed of an abscissa-parallel section with the points AB, an increasing section with the points BCD and an abscissa-parallel section with the points DE. If the internal combustion engine 1 z. B. operated at full load, a first proportional coefficient kpl of kpSTAT is assigned to the second actual rail pressure pCR2 (IST) of 1800 bar via the first characteristic curve 15. The second actual rail pressure pCR2 (IST) increases z. B. due to load shedding, an increased first proportional coefficient kpl of kpDYN is calculated in the dynamic range DYN over the first characteristic 15, point C. A higher first proportional coefficient kpl causes an increase in the P component of the high-pressure regulator 9 and thus a reduction in the manipulated variable or reduction in the throttle cross section of the suction throttle. A second characteristic curve 16 is shown in FIG. A second lead time Tvl is assigned to the second actual rail pressure pCR2 (IST) via the second characteristic curve 16. The second characteristic curve 16 corresponds to the course of the first characteristic curve 15. The second actual rail pressure pCR2 (IST) is plotted in bar on the abscissa. The first lead time Tvl is plotted on the ordinate as the output variable. The second characteristic curve 16 comprises a stationary area STAT and a dynamic area DYN. The stationary area ends at a pressure value of 1800 bar. This corresponds to the maximum stationary rail pressure at full load. The dynamic range DYN starts at a pressure value of 1820 bar. To increase safety, a tolerance band TB of 20 bar is provided between the stationary and dynamic range. The second characteristic curve 16 is composed of an abscissa-parallel section with the points AB, an increasing section with the points BCD and an abscissa-parallel section with the points DE. If the internal combustion engine 1 z. B. operated at full load, the second actual rail pressure pCR2 (IST) of 1800 bar is assigned a first lead time Tvl of TvSTAT via the second characteristic curve 16. The second actual rail pressure pCR2 (IST) increases z. B. due to load shedding, an increased first lead time Tvl of TvDYN is calculated in the dynamic range DYN over the second characteristic 16, point C. A higher first lead time Tvl causes an increase in the D component of the high-pressure regulator 9 and thus a decrease the manipulated variable or reduction of the throttle cross section of the suction throttle.

FIG. 5 shows the high-pressure control circuit for controlling the rail pressure pCR in a second embodiment. This embodiment differs from the illustration according to FIG. 2 in that the output variable of the second filter 14, here the second actual rail pressure pCR2 (IST), is at one Summation point B is subtracted from the target rail pressure pCR (SL). The result corresponds to a second control deviation dR2. This is led to the high pressure regulator 9. In this embodiment, the P, D and DTl portion of the high pressure regulator 9 is largely determined by the second actual rail pressure pCR2 (IST) via the second control deviation dR2. Figures 6 to 9 correspond to this.

FIG. 6 shows a block diagram for calculating a proportional coefficient kp. 6 comprises as essential elements the first characteristic curve 15 for calculating the first proportional coefficient kpl and a third characteristic curve 17 for calculating a second proportional coefficient kp2. The first proportional coefficient kpl is calculated in accordance with the description of FIG. 3. The input variable of the third characteristic curve 17 corresponds to the second control deviation dR2. The output variable corresponds to the second proportional coefficient kp2. Values of the second control deviation dR2 in the positive / negative direction are plotted on the abscissa of the third characteristic curve 17. The ordinate corresponds to the second proportional coefficient kp2. A first limit value GW1 and a second limit value GW2 are shown on the abscissa. If the second control deviation dR2 is very large, kp2 is limited to a value GW3. A negative control deviation is present when the second actual rail pressure pCR2 (IST) becomes greater than the target rail pressure pCR (SL). In the case of very large positive second control deviations dR2, the second proportional coefficient kp2 is limited to the value GW4. In the range between the first limit value GW1 and the second limit value GW2, the second proportional coefficient kp2 is set to the value zero. It is clear from the third characteristic curve 17 that, in the case of a steady state, ie the second control deviation dR2 is almost zero, the second proportional coefficient kp2 has the value zero. At a summation point C the first proportional coefficient kpl, the second proportional coefficient kp2 and a third proportional coefficient kp3 are added. The third proportional coefficient kp3 can either be constant or can be calculated as a function of a target torque and / or the engine speed nMOT. The sum of the three proportional coefficients corresponds to the proportional coefficient kp.

FIG. 7 shows a block diagram for calculating the lead time Tv. As essential elements, FIG. 7 comprises the second characteristic curve 16 for calculating the first retention time Tvl as a function of the second actual rail pressure pCR2 (IST) and a fourth characteristic curve 18 for calculating a second retention time Tv2. The first derivative time Tvl is calculated analogously to the description of FIG. 4. The input variable of the fourth characteristic curve 18 is the second control deviation dR2. The output variable of the characteristic curve 18 corresponds to the second lead time Tv2. Values of the second control deviation dR2 in the positive / negative direction are plotted on the abscissa. The ordinate corresponds to the second lead time Tv2. A first limit value GW1 and a second limit value GW2 are shown on the abscissa. In the case of very large negative values of the second control deviation dR2, the second lead time Tv2 is limited to a value GW3. In the case of very large positive second control deviations dR2, the second lead time Tv2 is limited to the value GW4. In the range between the first limit value GW1 and the second limit value GW2, the second lead time Tv2 is set to the value zero. From the fourth characteristic curve 18 it is clear that in a steady state, i.e. H. the second control deviation dR2 is almost zero, the second derivative time Tv2 has the value zero.

At a summation point C, the first lead time Tvl, the second lead time Tv2 and a third lead time Tv3 are added. The third lead time Tv3 can either be constant or depending on a target torque and / or the engine speed nMOT can be calculated. The result corresponds to the output variable Tv.

FIG. 8 shows a block circuit diagram for calculating the volume flow qV, that is to say the manipulated variable of the high-pressure controller 9. The internal structure of the high-pressure controller 9 is shown here. This contains three function blocks for calculating the controller shares. These are a P component 20, an I component 21 and a DTI component 22. A proportional component qV (P) of the volume flow qV is calculated via the P component 20 as a function of the first control deviation dRl and second control deviation dR2. An integrating component qV (I) of the volume flow qV is calculated via the I component 21 as a function of the first control deviation dRI. The DTl component qV (DTl) of the volume flow qV is calculated via the DTl component 22. The volume flow qV is determined via a summation 23 from the summands of the P, I and DTl component. There are two configurations for calculating the DTl component. In the first embodiment, the second control deviation dR2 is only used to calculate the derivative action time Tv, corresponding to FIG. 7. The input variable of the DTI algorithm is the first control deviation dRl. In the second embodiment, the input variable of the DT1 algorithm is also determined from the second control deviation dR2. For this purpose, FIG. 9 shows a diagram 19 of the DTI component qV (DTl) in the case of a sudden change in the input variable dR2. Time t is plotted on the abscissa. The ordinate corresponds to the DTI component qV (DTl). Two limit values GW1 and GW2 are shown in the diagram. The DTl component is deactivated when the second control deviation dR2 becomes smaller than the first limit value GWl, ie the signal qV (DTl) then has a value of zero. The DTI component is activated when the second control deviation dR2 becomes larger than the second limit value GW2. The limit value GW2 has the effect that, in the event of dynamic changes in state, that is to say a large positive or negative second control deviation dR2, the DTl component is used to calculate the volume flow qV miteingeht. In the case of stationary states, ie the second control deviation dR2 is almost zero, the volume flow qV is determined exclusively from the P component 20 and the I component 21.

reference numeral

Internal combustion engine fuel tank high pressure pump with suction throttle electronic control unit (ADEC) rail pressure sensor rail injector pressure relief valve high pressure regulator limitation function block controlled system first filter second filter first characteristic curve (kpl) second characteristic curve (Tvl) third characteristic curve (kp2) fourth characteristic curve (Tv2) Diagram P-part I-part DTl-part summation

Claims

claims

Process for regulating a pressure

Accumulator injection system of an internal combustion engine (1), in which a first actual rail pressure (pCRl (IΞT)) is determined via a first filter (13) from the measured rail pressure (pCR), a first control deviation (dRl) from a target rail pressure (pCR (SL)) and the first actual rail pressure (pCRl (IST)) is calculated and a volume flow (qV) is set via a high-pressure controller (9) as a manipulated variable as a function of the first control deviation (dRl), characterized in that a second actual rail pressure (pCR2 (IST)) is determined via a second filter (14) from the measured rail pressure (pCR) and the second actual rail pressure (pCR2 (IST)) is decisive for the calculation of the controller components of the high pressure Controller (9) is set.

Method according to Claim 1, characterized in that a first proportional coefficient (cpl) for determining a P component (20) of the high-pressure regulator (9) via a first characteristic curve (15) as a function of the second actual

Rail pressure (pCR2 (IST)) is calculated.

A method according to claim 1 and 2, characterized in that in addition a first lead time (Tvl) for

Determination of a D component and a DTI component (22) of the

High-pressure controller (9) over a second characteristic (16) in

Dependency of the second actual rail pressure (pCR2 (IST)) is calculated.

4. The method according to claim 3, characterized in that the first characteristic curve (15) has a stationary area (STAT) and dynamic area (DYN), wherein in the stationary area (STAT) regardless of the value of the second actual rail pressure (pCR2 (IST )) a constant first proportional coefficient (kpl) is calculated (kpl = const) and in the dynamic range (DYN) the first proportional coefficient (kpl) is calculated according to an increasing function.

5. The method according to claim 3, characterized in that the second characteristic curve (16) has a stationary area (STAT) and dynamic area (DYN), wherein in the stationary area (STAT) regardless of the value of the second actual rail pressure (pCR2 (IST )) a constant first lead time (Tvl) is calculated (Tvl = const) and in the dynamic range (DYN) the first lead time (Tvl) is calculated according to an increasing function.

6. The method according to claim 1, characterized in that a second control deviation (dR2) is calculated from the target rail pressure (pCR (SL)) and the second actual rail pressure (pCR2 (IST)) and the P component (20) , D component and the DTl component (22) of the high-pressure controller (9) are determined as a function of the second control deviation (dR2).

7. The method according to claim 6, characterized in that a proportional coefficient (kp) for determining the P portion of the high pressure regulator (9) from the sum of first (kpl), a second (kp2) and a third proportional coefficient (kp3) is calculated.

8. The method according to claim 7, characterized in that the first proportional coefficient (kpl) is calculated according to claim 2.

9. The method according to claim 7, characterized in that the second proportional coefficient (kp2) is calculated as a function of the second control deviation (dR2) via a third characteristic curve (17).

10. The method according to claim 7, characterized in that the third proportional coefficient (kp3) is either constant (kp3 = const) or calculated depending on a target torque and / or an engine speed.

11. The method according to claim 6, characterized in that a lead time (Tv) for determining the D component of the high pressure regulator (9) is calculated from the sum of the first, a second and a third lead time (Tvl, Tv2, Tv3) ,

12. The method according to claim 11, characterized in that the first lead time (Tvl) is calculated according to claim 3.

13. The method according to claim 11, characterized in that the second lead time (Tv2) is determined as a function of the second control deviation (dR2) via a fourth characteristic curve (18).

14. The method according to claim 11, characterized in that the third lead time (Tv3) is either constant (Tv3 = const) or is calculated as a function of a target torque and / or an engine speed.

15. The method according to claim 6, characterized in that the DTl portion (22) of the high-pressure controller (9) is switched on when a second limit value (GW2) is exceeded, and is switched off when a first limit value (GWl) is undershot becomes.

16. The method according to any one of the preceding claims, characterized in that the second filter (14) has a smaller time constant and a smaller phase delay than the first filter (13).