GB2478647A

GB2478647A - Method of Controlling the Combustion of an Internal Combustion Engine

Info

Publication number: GB2478647A
Application number: GB1103988A
Authority: GB
Inventors: Alessandro Catanese; Gerhard Landsmann; Andrea Emilio Catania
Original assignee: GM Global Technology Operations LLC
Current assignee: GM Global Technology Operations LLC
Priority date: 2010-03-09
Filing date: 2011-03-09
Publication date: 2011-09-14
Anticipated expiration: 2031-03-09
Also published as: GB201003856D0; GB2481364A; GB2478647B; GB201103988D0

Abstract

A method for controlling the combustion in an internal combustion engine comprising; calculating the start time of the next injection pulse and the amount of fuel to be injected by that pulse, calculating the heat release rate of the next injection pulse, and correcting the time of the next injection pulse and the amount of fuel to be injected by that pulse on the basis of the calculated heat release rate.Even if having the same start of injection and the same injection flow, combustion events within the same engine cylinder can develop with very different speed and characteristics. This problem can be monitored and controlled by adding cylinder pressure sensors to the engine control architecture, however the sensors only supply information after the combustion has happened, so are suitable for closed loop corrections and not for predictive control methods. The invention estimates various indexes of the combustion phasing by means of existing information prior to the combustion event happening, therefore cylinder pressure sensors are not required.

Description

Method of controlling the oanbustion of an internal ccxnbustion engine TECHNICAL FTh'TTI The present invention relates to a method of controlling the conibus-tion of an internal combustion engine, i.e. a compression ignition internal combustion engine of a motor vehicle, typically a Diesel en-gine, during an engine cycle.

In current Diesel engines, the fuel is injected in the combustion chamber by means of a plurality of injection pulses per engine cycle according to a multi-injection pattern, typically by means of at least a pilot injection pulse and a following main injection pulse.

The combustion phasing is controlled acting on the start of injection (SOl) of said injection pulses. Current injectors have reached a very * .* S..

* 20 good quality and permit a high repeatability of the needle opening :.. actuation across production dispersion and mileage. Nonetheless, the start of injection is not univocally related to the combustion phas- ing, indeed according to engine operating conditions and environmen- *:*. tal conditions, even with the same start of injection (SOl) and the same injection flow, the combustion could develop with very different speed and characteristics. This variability is further increased by the fact that modern Diesel engines are using increasingly higher rate of exhaust gas recirculation, which further increase the time delay between the start of injection and the start of combustion.

To monitor the combustion phasing, cylinder pressure sensors have been added to the engine control architecture. These sensors allow to measure the development of the combustion process and to calculate some more relevant indexes based on the heat released by the cornbus-tion. The drawback of this approach is in the very high cost of this sensors and in the fact that they give information on the combustion phasing after the combustion has happened. Therefore they cannot be used for feed-forward control of the injection and combustion but on-ly for closed loop corrections.

In order to overcome this drawback, it is advisable to estimate one or more relevant indexes of the combustion phasing by means of exist-ing information (no need to add cylinder pressure sensors), prior than combustion happens.

Several models exist in literature to predict the combustion event S.....

and the heat release, by way of exarrle from Rolf Egnell "A Simple Approach to Study the Relation Between Fuel Rate, Heat Release Rate and NOx-Formation in Diesel Engines", SAE Technical Paper 1999-01- . : 3548, published on October 1999, pages 4-6; or from Fabrizio Ponti, *:*.; Enrico Corti, Gabriele Serra and Matteo De Cesare "Corrmon Rail Multi-Jet Diesel Engine Combustion Model Development for Control Purposes" SAE Technical Paper 2007-01-0383, published on Ipril 2007, pages 2-4; or from Franz G. Chrnela and Gerhard C. Orthaber "Rate of Heat Release Prediction for Direct Injection Diesel Engines Based on Purely Mixing Controlled Combustion", SAE Technical Paper 1999-01-0186, published on March 1999, pages 2-5; or from Claes Ericson, Bjorn Westerberg, Magnus Pndersson and Roif Egnell "Modelling Diesel Engine Combustion and NOx Formation for Model Based Control and Simulation of Engine and Exhaust Aftertreatment Systems", SAE Technical Paper 2006-01- 0687, published on april 2006, pages 4-5.

The major limitations of these existing models are either in a too high complexity, leading to a not feasible implementation in real time control system applications, or in a too rough simplification, leading to an inaccuracy of the results outside very strict applica-tion boundaries.

In view of the above, it is an object of the present invention to provide a model based combustion control strategy capable to control fuel injections with enough accuracy and acceptable computational re-quirements, such as to be effectively usable in a control system of * . an engine. *

S.....

* * 20 Another object of the present invention is to achieve the above men- tioned purpose with a simple, rational and rather inexpensive solu- :::. tion. 0

DISOSURE

These and/or other objects are attained by the characteristics of the embodiments of the invention as reported in independent claims. The dependent claims recite preferred and/or especially advantageous fea-tures of the embodiments of the invention.

n embodiment of the invention provides a method of controlling the combustion of an internal combustion engine, said engine injecting during an engine cycle a quantity of fuel through at least one injec-tion pulse i. The method firstly comprises step of calculating the time t of the next injection pulse and the amount of fuel to be in-jected by that injection pulse. Then, the heat release rate dQ(t) of said next injection pulse i is calculated according to the following equation: dQ, (t) = k. LQFUEL, (t -. Q(t)] wherein k is an index representing the speed of the combustion process during said at least an injection pulse i, Q(t) is a total heat released at the time t, QH,j(t-rj) is the energy of the fuel in-jected at the time t-r by said at least an injection pulse i, r1 is an index representing the delay between the time at which a certain * fuel quantity of said at least an injection pulse i is injected and * S * the time at which it actually burns. Then, the method performs a cor-

S S * S

:.. 20 rection, on the basis of said calculated heat release rate, of the * time t of the next injection pulse and of the amount of fuel to be injected by that injection pulse. In the simplest way this can be S* done by replacing the values (of time and amount of fuel) calculated in the first place by the values derived by solving the equation (de-rived values). In a more realistic scenario, which takes more factors of the engine into account, a difference between the calculated val-ues and the derived values are used to correct the calculated values by shifting them towards the derived values to a certain extent. Fi- nally, the method initiates the next injection pulse with the cor-rected value of time t and the corrected amount of fuel.

This method allows a reliable estimation of the heat release rate with the aid of a simple equation, which requires limited corriputa-tional efforts and less memory space. It avoids a rigid injection pattern and allows flexibly adapting the injection pattern in real time to optiinise fuel efficiency, reduce fuel consumption and thus emissions.

It is to be noted that the released heat is calculated before this amount of fuel is actually injected, such that the calculated heat release is used to predictively correct the amount of fuel. This pre- dictive approach helps avoids the use of sensors to monitor the com-bustion itself, thus reducing the complexity of the engine and costs associated with it.

According to an embodiment of the invention, the energy of the in-* ** *** * 20 jected fuel at the time t-rj is calculated according to the following equation: QFuEL,i(t-v!) H frn(t_r1)dt soI' ** * * * * * ** wherein H is the lower heating value of the fuel, SOIL is the start of injection and m1 is the fuel injection rate of said at least an injection pulse i.

This embodiment has the advantage of providing a simple way to calcu- late the energy of the injected fuel, which further reduces the com-putational and memory efforts.

The index r can be determined implementing various strategies that are currently available in literature, by way of example from H. Kim and N. Sung "Combustion and Emission Modeling for a Direct Injection Diesel Engine", SAE Technical paper 2004-01-0104, published on 2004, page 5.

Nonetheless, a preferred embodiment of the invention provides for de- termining the index r on the base of at least a parameter representa- tive of the energy released by the combustion and a parameter repre-sentative of the air density in the combustion chamber, this latter being preferably considered at the time in which the piston is at the Top Death Center (T1X).

This embodiment of the invention provides a feasible strategy for de-termining the index r, because it has been found that an effective * ** *** * . 20 correlation exists between the heat released by the combustion, the air density in the combustion chamber and the ignition delay of an injected fuel quantity. ** * * * *

According to an aspect of this embodiment, the index r is thus deter-mined through an empirically determined relationship correlating the index r1 to at least the above mentioned combustion energy and air density related parameters.

This relationship can be advantageously determined through an experi- mental activity and then implemented as a subroutine, thereby allow-ing the determination of the index r with a minimum of computational effort.

Since different kinds of injection pulse, such as for example pilot injection pulse or main injection pulse, may be characterized by a different ignition delays, also the empirical relationship between r2, combustion energy and air density could be different.

As a consequence, the above mentiohed empirical relationship is pre-ferably determined for each injection pulse i independently, with the aid of a dedicated experimental activity focused on said specific in-jection pulse i, thereby allowing a more accurate estimation of the heat release rate caused by the fuel injection.

In order to simplify the experimental activity and reduce the compu-tational and memory efforts, the index r can determined through an * ***.* empirically determined relationship correlating the index r to at least a single factor, which is a function of both the above men-tioned combustion energy and air density related parameters. * * S

*:.*:3 According to an embodiment of the invention, the index k is deter-mined on the base of at least a parameter representing the engine speed and a parameter representing the quantity of fuel to be in-jected by said at least an injection pulse i.

This embodiment of the invention provides a reliable strategy for de-termining the index k2, because it has been experimentally found that an effective correlation exists between combustion speed, engine speed and fuel injected quantity.

According to an aspect of this embodiment, the index k is thus deter-mined through an empirically determined relationship correlating the index k1 to at least the above mentioned engine speed and injected fuel quantity related parameters.

This relationship can be advantageously determined through an experi- mental activity and then implemented in a subroutine, thereby allow-ing the determinatIon of the index k with a minimum of computational effort.

Also in this case, the empirical relationship between k1, engine speed and fuel injected quantity could be different for different kinds of injection pulse, so that this empirical relationship is preferably :: 20 determined for each injection pulse i independently, by means of a dedicated experimental activity specially focused on said specific injection pulse i. *. . * . * SI

** * According to an embodiment of the invention, the method comprises the

I * Sc

further step of time integrating the calculated heat release rate dQ(t), so as to calculate a total heat Q released by said at least an injection pulse i.

In this way, the method advantageously allows a reliable estimation not only of the heat release rate, but also of the heat that is to-tally released during a combustion process.

According to another embodiment of the invention, when the fuel is injected through a plurality of injection pulses, the method provides for calculating the heat release rate dQ1(t) of each injection pulse separately.

Thanks to this solution, it has been found that the estimation of the heat release rate during the combustion process results more accu-rate.

According to an aspect of this embodiment, the method comprises the step of adding the heat release rates dQ(t) calculated for each in-jection pulse, so as to calculate a cumulative heat release rate **.*..

* : dQ(t).

S..... * .

:. 20 In this way, the cumulative heat release rate results a reliable in- dex of the combustion process, which takes into account the contribu- tions of all injection pulses. S. * S * S S *s

According to another aspect of this embodiment, the method comprises the further step of time integrating the calculated cumulative heat release rate dQ(t), so as to calculate a cumulative heat Q totally released by said quantity of injected fuel.

In this way, the cumulative heat results still another reliable index of the combustion process, which represents the heat that is totally released during the combustion process with the contributions of all injection pulses.

The method according to the invention can be carried out with the help of a computer program comprising a program-code for carrying out all the steps of the method described above, and in the form of a computer program product comprising the computer program.

The computer program product can be embodied as an internal combus-tion engine comprising an engine control unit (ECU), a data carrier associated to the ECU, and the computer program stored in the data carrier, so that, when the ECU executes the computer program, all the steps of the method described above are carried out.

I

* * The method can be also embodied as an electromagnetic signal, said **S*..

signal being modulated to carry a sequence of data bits which * * 20 represent a computer program to carry out all steps of the method.

DESIPTI OF THE DRAWINGS

The present invention will now be described, by way of example, with reference to the accompanying drawing.

Figure 1 is a flowchart representing a combustion estimating method according to an embodiment of the invention.

DETAILED DESCRIPTI(}T An embodiment of the present invention is hereinafter disclosed with reference to a conventional Diesel engine of a motor vehicle (not shown), which generally comprises a plurality of cylinders that indi-vidually defines a combustion chamber.

The combustion chamber is closed at the bottom by a reciprocating piston, whose movement is caused by the combustion process of a quan-tity of fuel that is injected into the combustion chamber during each engine cycle.

The present embodiment of the invention provides a method for esti-mating the dynamic of said combustion process.

During each engine cycle, the fuel is conventionally injected into the combustion chamber by means of a plurality of injection pulses c..

* that follows one another, according to a predetermined multi-S..... * .

injection pattern S. *S * . S * * 20 The fuel quantity that actually burns into the combustion chamber is usually injected by means of at least a pilot injection pulse and a *.5. following main injection pulse.

However, the present embodiment of the invention does not exclude that the total burning fuel quantity may be injected through a single injection pulse, or through more than two injection pulses of the multi-injection pattern.

Accordingly, the generic injection pulse of burning fuel is hereinaf-ter indicated with 1, wherein i can vary from 1 to n, wherein 1 represents the first injection pulse of burning fuel and n represents the last injection pulse of burning fuel, and wherein n can be equal or greater than one.

The combustion estimating method provides a first routine, represented by the block 10 of figure 1, for separately calculating the heat release rate dQ1(t) caused by each injection pulse i at the generic time t.

Each heat release rate dQ1(t) is calculated according to the following equation: dQ1(t) = k, LQFUEL,I (t -r,) -Q(t)j wherein k1 is an index representing the speed of the combustion process calibrated for the 1th injection pulse, Q(t) is the total heat released up to the time t, Q,(t-r) is the energy of the fuel in-jected by the 1th injection pulse at the time t-r, tj is an index . : representing the delay between the time at which a certain fuel quan- **.: tity is injected by the th injection pulse and the time at which it actually burns.

The energy Q,(t-r1) can be calculated according to the following equation: QIuEL,!(t-ri) = H, Jrn1(t_r1)dt soil wherein H is the lower heating value of the fuel, SOIL is the start of injection of the 1th injection pulse, and m1 is the fuel injection rate of the 1th injection pulse.

In the present embodiment, each index r is determined on the base of a single factor, also referred as Waiting Factor WF, which is a func-tion of two key parameters that affect the ignition delay, namely a parameter representing the energy released by the combustion during an engine cycle and the air density prr,-in the combustion chamber when the piston is at the TEC.

The air density p can be estimated according to the following equa-tion: -i,, (ic

PTDC -I TI I1,,, a

* wherein Pm is the pressure of the air in the intake manifold of the * S....

* engine, R is the specific gas constant, Tm is the temperature of the air in the intake manifold, V is the combustion chamber volume when the piston is at the Bottom Death Center, and V is the combustion *. 20 chamber volume when the piston is at the TDC. S. I * * a * *I

The combustion energy related parameter can be either an Indicated Mean Effective Pressure (IMEP) or equivalently a fuel injected quan-tity qj,j to be injected by the th injection pulse.

In view of the above, WF can be calculated according to the following equation: WFC.1PT+D.I1M PTDCO) LIMEP0 or equivalently / ( WF=C.1 PTDC 1 +D.I PTDCO) q,10 wherein a, b, C and D are empirical constant dependent by the specif-ic engine; p, IMEP0 and qinj,jo are respectively the air density at the TDC, the indicated mean effective pressure and the fuel injected quantity in a reference condition 0.

By way of example, for a turbocharged CIDI (Compression Ignition Di- rect Injection) engine having a displacement of 2 liters and a. Com-pression Ratio of 15.5, the following parameter can be assumed: a=2, b=2, C=0.5, D=0.3, p=35kg/m3, IMEP0=25bar and qwi,jcFr8OnlTIrn3. : 15 * .

*: Returning now to the combustion estimating method, each index r1 is determined through a dedicated subroutine that correlates the index r * * * * S to the waiting factor WF. ** *

For the sake of simplicity, these dedicated subroutines are globally * represented by a single block 11 in figure 1.

The relationship between WF and r implemented in each subroutine is empirically determined through an experimental activity focused on the correspondent th injection pulse.

This experimental activity generally provides for operating a test Diesel engine, having the same characteristic of that to which the combustion estimating method is directed, in a plurality of different conditions, mainly at a plurality of different values of WF (i.e. at a plurality of different values of the combustion energy related pa-rarneter and of the air density); for measuring the heat release rate dQ due to the 1th injection pulse in each condition with the aid of a proper sensor (i.e. a cylinder pressure sensor); for calculating the index r in each condition; and for interpolating the calculated in-dexes Ti, in order to determine the empirical relationship between WF and r1, which is finally implemented in the subroutine.

As a matter of fact, this experimental activity must be performed for each kind of injection pulse i separately, by way of example for both the pilot injection pulse and the main injection pulse, in order to obtain a plurality of subroutine, each of which implements a respec-S..

* tive relationship between WF and the index r1, which is specifically * *. *.* calibrated for the correspondent injection pulse i.

By way of example, an experimental activity of this kind, performed for the main injection pulse of the above mentioned turbocharged CIDI *,.: (Compression Ignition Direct Injection) engine having a displacement of 2 liters and a Compression Ratio of 15.5, has returned the follow-ing empirical relationship: rW(,fl[iWI=KJWF if WF<0.6 and Vrnajn[iS1 120 if WF'>0.6 wherein r is the index referred to the main injection pulse, K0=1801.3, K116443, 1(2=68524, K2=153979, K4=194421, K5=129332 and K6=35118.

In the present embodiment, each index k is determined on the base of the engine speed and the quantity of fuel qinj,j to be injected by the 1th injection pulse.

In greater detail, each index k is determined through a dedicated subroutine, which is calibrated for the correspondent th injection pulse, and which correlates the index k to the engine speed co and the quantity of fuel qj3,j.

For the sake of simplicity, these dedicated subroutines are globally *: represented by a single block 12 in figure 1.

Each dedicated subroutine implements a relationship between co, qj,j and k, which is empirically determined through an experimental activ- ::::; ity focused on the correspondent th injection pulse.

This experimental activity generally comprises the steps of: operat-ing a test Diesel engine, having the same characteristic of that to which the combustion estimating method is directed, in a plurality of different conditions, mainly at a plurality of different values of the quantity of fuel qjnj,j and of the engine speed co; measuring the heat release rate dQ due to the 1th injection pulse in each condition with the aid of a proper sensor (i.e. a cylinder pressure sensor); calculating the index k in each condition; and interpolating the cal-culated indexes k1, in order to determine the empirical relationship between w, qinj,i and k, which is finally implemented in the dedicated subroutine.

As a matter of fact, this experimental activity must be performed for each kind of injection pulse i separately, by way of example for both the pilot injection pulse and the main injection pulse, in order to obtain a plurality of subroutines, each of which implements a rela-tionship between w, qj,j and k1 that is specifically calibrated on the correspondent injection pulse i.

By way of example, an experimental activity, performed for the main injection pulse of the above mentioned turbocharged CIDI (Compression Ignition Direct Injection) engine having a displacement of 2 liters S.....

* and a Compression Ratio of 15.5, has returned the following empirical *.*...

relationship: * 20 k,,,0,=O.4819*w+848.45 if qjnj,n<2O mm3; : k11111,O.2992*w+532.17 if qjnj,n>2O nun3; (O.4819*&+848.45)-(O.2992.an-532.17) *. . k -20)+O.4819.w+848.45 * * * main,?y,maAfl * ** if 20 mm3.

After the individual heat release rate dQ(t) have been calculated, the combustion estimating method provides a second routine, represented by the block 13 of figure 1, for calculating a cumulative heat release rate dQ(t) caused by all the injection pulses at the ge-neric time t.

The cumulative heat release rate dQ(t) is calculated according to the following equation: dQ(t) = >dQ1(t) wherein dQ(t) is the heat release rate caused by each injection pulse i at the time t.

After the cumulative heat release rate dQ(t) has been calculated, the combustion estimating method provides a third routine, represented by the block 14 in figure 1, for calculating a cumulative heat Q global-ly released by the injected burning fuel.

The cumulative heat Q is calculated by time integrating the cumula-* * tive heat release rate dQ(t) according to the following equation: * ***.* Q= JdQ(t)dt or equivalently according to the following equation: * wherein Q is the total heat released by the th injection pulse and is calculated by time integrating dQ(t) as follow: = JdQ,Q)dt While at least one exemplary embodiment has been presented in the foregoing surrinary and detailed description, it should be appreciated that a vast number of variations exist. It should also be appreciated that the exemplary embodiment or exemplary embodiments are only exam- ples, and are not intended to limit the scope, applicability, or con- figuration in any way. Rather, the forgoing summary and detailed de-scription will provide those skilled in the art with a convenient road map for implementing at least one exemplary embodiment, it being understood that various changes may be made in the function and ar-rangement of elements described in an exemplary embodiment without departing from the scope as set forth in the appended claims and in their legal equivalents.

I

S..... * . a

* *.... * S *. S. * * S * S I. * * . . * SS * . * S I **

EFER NUMS

First routine ii ii Subroutine 12 Ki subroutine 13 Second routine 14 Third routine ii Ignition delay index Engine speed related parameter qinj,i Injected fuel quantity ki Combustion speed index dQi(t) Heat release rate dQ(t) Cumulative heat release rate Q Cumulative heat released WF Waiting factor * S

S

* 54.s. * *. . a. * U * S. a * S * * o.

SS S * * S * .#

Claims

1. Method of controlling the combustion of an internal combustion engine, said engine injecting during an engine cycle a quantity of fuel through at least one injection pulse i, the method corn-prising the steps of a) calculating the start time (SOl) of the next injection pulse and the amount of fuel to be injected by that injection pulse, d) calculating the heat release rate dQ(t) of said next injec-tion pulse i according to the following equation: dQ1(t) k1 [QFUEL,(t-r)-Q(t)i wherein k is an index representing the speed of the combustion process during said at least an injection pulse i, Q(t) is a to-tal heat released at a time t, Q,(t-r) is the energy of the fuel injected at the time t-r by said at least an injection pulse i, r is an index representing the delay between the time at which a certain fuel quantity of said at least an injection pulse i is injected and the time at which it actually burns, c) correcting, on the basis of said calculated heat release rate, :1..: the time t of the next injection pulse and the amount of fuel to be injected by that injection pulse.. : d) initiating the next injection pulse with the corrected value of time t and the corrected amount of fuel.

2. Method according to claim 1, wherein the energy of the injected fuel at the time t-r is calculated according to the following equation: QFUEL,1 (t -r1) = H,, Jrn, (t -r)dt soil wherein H is the lower heating value of the fuel, SOIL is the start of injection and m is the fuel injection rate of said at least an injection pulse i.

3. Method according to claim 1, wherein the index r is determined on the base of at least a parameter (IMEP, qj,j) representing the energy released by the combustion and a parameter (P2w) representing the air density in the combustion chamber.

4. Method according to claim 3, wherein the index r is determined through an empirically determined relationship correlating the index r to at least the combustion energy related parameter (IM-EP, qinj,i) and air density related parameter (p2w).

5. Method according to claim 3, wherein the index r1 is determined through an empirically determined relationship correlating the index r to at least a single factor (WF, which Is a function of both the combustion energy related parameter (IMEP, qj,i) and air density related parameter (p2w).

6. Method according to claim 1, wherein the index k is determined SISS.* : on the base of at least of a parameter representing the engine S * speed () and a parameter representing the quantity of fuel *. S S. S (qj,j) to be injected by said at least an injection pulse i.

7. Method according to claim 5, wherein the index k is determined through an empirically determined relationship correlating the index k to at least the engine speed related parameter () and injected fuel quantity related parameter 8. Method according to claim 1, wherein the estimating strategy com-prises the step of time integrating the calculated heat release rate dQ(t), so as to calculate a total heat Q. released by said at least an injection pulse i.9. Method according to claim 1, wherein said quantity of fuel is in-jected through a plurality of injection pulses, and wherein the estimating strategy provides for calculating the heat release rate dQ1(t) of each injection pulse separately.10. Method according to claim 9, wherein the estimating strategy com- prises the step of adding the heat release rates dQ(t) calcu-lated for each injection pulse, so as to calculate a cumulative heat release rate dQ(t).11. Method according to claim 10, wherein the estimating strategy comprises the step of time integrating the calculated cumulative heat release rate dQ(t), so as to calculate a cumulative heat Q totally released by said quantity of injected fuel.12. Computer program comprising a computer-code for carrying out a method according to any of the preceding claims.13. Computer program product on which the computer program according to claim 12 is stored.14. Compression ignition internal combustion engine comprising an ECU, a data carrier (14) associated to the ECU, and a computer 20 program according to claim 12 stored in the data carrier.15. J\n electromagnetic signal modulated as a carrier for a sequence of data bits representing the computer program according to claim 12.

S. I * * . * S. S. * * * . *.