Real-Time Determination of a Fresh-Air Mass in a Cylinder

Abstract

Please substitute the new Abstract submitted herewith for the original Abstract: A method for determining a fresh-air-mass parameter in a cylinder of an internal combustion engine in a motor vehicle includes identifying a cylinder which is at the end of an intake stroke or at the beginning of a compression stroke during operation of the motor vehicle, determining a diagnosis time window which lasts for a period of time after closing of the inlet valves of the identified cylinder within a region of low torque of the internal combustion engine, ascertaining a development in the speed of the internal combustion engine during the diagnosis time window, and determining the fresh-air-mass parameter in the identified cylinder.

Claims

1-15. (canceled)

16. A method for determining a fresh-air mass parameter in a cylinder of an internal combustion engine in a motor vehicle, the method comprising: identifying a cylinder which is at the end of an intake stroke or at the beginning of a compression stroke during driving operation of the motor vehicle; determining a diagnosis time window which lasts for a period of time after a closing the inlet valves of the identified cylinder within a low-torque range of the internal combustion engine; determining a development in the speed of the internal combustion engine during the diagnosis time window with a real-time-capable sampling quality; determining a simplified cylinder load parameter in the identified cylinder in dependence on the determined development in the speed; and determining the fresh-air mass parameter in the identified cylinder in dependence on the determined simplified cylinder load parameter.

17. The method according to claim 16, comprising: wherein the following variable quantities are determined in addition to the development in the speed in order to determine the simplified cylinder load parameter (rf*): determining a cylinder volume at a diagnosis time point which lies within the diagnosis time window in order to determine the simplified cylinder load parameter in addition to the development in the speed; and/or determining a reduced piston acceleration in the diagnosis time window in order to determine the simplified cylinder load parameter in addition to the development in the speed.

18. The method according to claim 17, comprising: using only constants to determine the simplified cylinder load parameter.

19. The method according to claim 17, comprising: determining a pressure characteristic for the identified cylinder in the diagnosis time window in dependence on the determined development in the speed and/or the determined reduced piston acceleration; and determining the simplified cylinder load parameter in dependence on the determined pressure characteristic and/or the determined cylinder volume.

20. The method according to claim 16, comprising: determining, before the other method steps, whether at least quasi steady-state operation or transient operation of the internal combustion engine is present.

21. The method according to claim 20, comprising: carrying out the method if and/or as long as it is determined that that a transient operation of the internal combustion engine is present.

22. The method according to claim 20, comprising: storing and/or using the determined residual gas fraction if and/or as long as it is determined that transient operation of the internal combustion engine is present.

23. The method according to claim 16, comprising: determining the fresh-air mass parameter in the identified cylinder: only on a basis of the determined, simplified cylinder load parameter, or additionally on a basis of a steady-state cylinder load parameter determined for steady-state operation and/or an offset prediction of the fresh-air mass parameter, which is made in dependence on a steady-state cylinder load parameter.

24. The method according to claim 16, wherein the fresh-air mass parameter determined for a particular operating cycle of the identified cylinder serves as a basis for a determination of a fuel injection quantity for the subsequent operating cycle of the cylinder or a subsequently firing cylinder.

25. The method according to claim 16, comprising: determining the development in the speed with a real-time-capable sampling quality.

26. An engine control unit for determining a fresh-air mass parameter in a cylinder of an internal combustion engine, the engine control unit configured to: identify a cylinder which is at the end of an intake stroke or at the beginning of a compression stroke during driving operation of the motor vehicle; determine a diagnosis time window which lasts for a period of time after a closing the inlet valves of the identified cylinder within a low-torque range of the internal combustion engine; determine a development in the speed of the internal combustion engine during the diagnosis time window with a real-time-capable sampling quality; determine a simplified cylinder load parameter in the identified cylinder in dependence on the determined development in the speed; determine the fresh-air mass parameter in the identified cylinder in dependence on the determined simplified cylinder load parameter; transfer values of the fresh-air mass parameter to at least one of a control component of the engine control unit for real-time control of functions of the internal combustion engine in dependence on the transferred values, or to a diagnostic component of the engine control unit for further onboard diagnostic functions.

27. The engine control unit according to claim 26, configured to: determine a fuel injection quantity for a specific operating cycle of a cylinder in dependence on a determined value of the fresh-air mass parameter with respect to a previous operating cycle of the identified cylinder or of a cylinder diagnosed directly before.

28. The engine control unit according to claim 27, configured to: evaluate a specific cylinder and use its previously determined air mass in a new operating cycle to determine the fresh-air mass parameter.

29. The engine control unit according to claim 27, configured to: evaluate a cylinder with regard to its air mass and the value of the fresh-air mass parameter determined at that time and/or determined last is transferred to the next firing cylinder for its feedforward control.

30. The engine control unit according to claim 26, comprising: a non-volatile memory, wherein the engine control unit is configured to store one or more values of the fresh-air mass parameter determined at one or different diagnosis time windows in the non-volatile memory.

31. The engine control unit according to claim 30, configured to transfer the values of the fresh-air mass parameter stored in the memory to an offboard computer for off-line diagnostic functions.

32. An internal combustion engine comprising: one or more cylinders; and the engine control unit according to claim 26.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

[0051] FIGS. 1A, 1B, and 1C show schematic views of an internal combustion engine with an engine control system according to an exemplary embodiment of the invention, wherein FIG. 1A shows the installation environment of the internal combustion engine,

[0052] FIG. 1B shows relevant parameters and FIG. 1C shows torque contributions at the crank mechanism of the internal combustion engine over time.

[0053] FIG. 2 shows a diagram with a graph of a development in the speed of an operating cycle of the internal combustion engine according to FIG. 1 and a representation of the cycles of the individual cylinders.

[0054] FIG. 3 shows an enlarged detail from the graph in FIG. 2.

[0055] FIG. 4 shows a diagram of a mass balancing in the selected cylinder.

[0056] FIGS. 5A, 5B, 5C, 5D, 5E and 5F illustrate the process of reducing the complex relationships of a residual gas mass and a temperature in the cylinder to a simple set of formulas that enables real-time calculation on an engine control unit.

[0057] FIG. 6 shows a method carried out in an exemplary embodiment.

DETAILED DESCRIPTION

[0058] In FIG. 1B, the internal combustion engine 1 is shown in a more detailed schematic view. The internal combustion engine 1 has cylinders Z1, Z2, Z3 and Z4, wherein all cylinders Z provide their torque contribution M to a crankshaft of a crank drive KT. The internal combustion engine 1 additionally comprises a control means 2 according to an exemplary embodiment of the invention, optionally comprising a computing unit 4 if the control means 2 is not formed as part of an engine control unit. The control means 2 further comprises a speed detection unit 6 and a cylinder pressure detection unit 7 for the reference pressures from the surrounding environment and air collector or crankcase. The control means 2 also has a cylinder volume determination unit and a cylinder temperature determination unit and can access measured values of all lambda sensors of the internal combustion engine 1.

[0059] FIG. 1B shows, inter alia, that, depending on the particular cylinder pressure p, each cylinder Z can cyclically apply a torque contribution M to the crankshaft drive KT. The totality of the torque contributions results in a time-variable speed n of a crankshaft of the crank drive KT.

[0060] The reference pressure p can be used by means of the pressure detection unit 7, the instantaneous speed n can be used by means of the speed detection unit 6, and the computing unit 4 can be used by the device 2.

[0061] FIG. 1C shows a graph of a torque development M.sub.ges with an exemplary torque curve 10 at the crankshaft drive KT during normal operation over the crank angle KW. It is evident that the torque contribution M comes alternately from different cylinders Z. A torque limit value 14 is drawn in the illustration and is set in particular arbitrarily and determines below which torque a torque contribution of a cylinder is considered insignificant, so that a low-torque range 12 in the sense of the invention then exists. Consequently, a low-torque range 12 in the sense of the invention can be identified if, at a certain time interval, the torque contributions of each cylinder are below the limit value 14.

[0062] In the embodiment of FIG. 1C, low-torque ranges 12 of slightly different lengths result. Within each of these low-torque ranges 12, in particular, a diagnosis time window 112 can be defined and may comprise the entire period of the low-torque range or a part thereof.

[0063] FIG. 2 shows a sketch of an exemplary graph 150 of a development in the speed 101 of a four-stroke cycle (=one operating cycle (ASP): top dead center charge change (LOT) .fwdarw.intake .fwdarw.bottom dead center (UT) .fwdarw.valve closure and compression .fwdarw.top dead center ignition (ZOT) .fwdarw.expansion .fwdarw.UT .fwdarw.exhaust) of the internal combustion engine 1.

[0064] The sequence diagram 150 shows the curve 101 of the engine speed n over an operating cycle (ASP) of a 4-cylinder gasoline engine. The ignition timing points (ZZP) and an example of a possible diagnosis time window 112 for the cylinder Z1 to be diagnosed in the compression phase are marked. Below this, the associated power strokes of the physical cylinders Z1-Z4 are shown.

[0065] This example of a four-cylinder engine shows which range 112 of the crank angle scale can be used for charge change diagnosis. The diagnosis time window 112 for the cylinder Z1 to be diagnosed is in the compression phase, i.e. when the intake phase has already been completed and there is also a low-torque range (cf. limit value 14 in FIG. 1C).

[0066] In particular, the diagnosis time window 112 must be selected so that the last cylinder performing work no longer achieves acceleration of the crankshaft and the next cylinder performing work has not yet fired.

[0067] In the exemplary embodiment, the diagnosis time window comprises a time interval in which the inlet valves of the cylinder Z1 to be diagnosed are closed again after the intake of the charge air or the combustion mixture, and in addition a low-torque range of the internal combustion engine 1 is present. The limits depend on an applied engine operating point and can be flexibly adapted thereto. Dynamic adaptation of the limits of the diagnosis time window 112 is also possible for dynamic driving operation in dependence on boundary conditions such as an ignition angle and the cylinder pressure curve.

[0068] In the exemplary embodiment, therefore, the diagnosis time window 112 is determined to be 660 KW to 690 KW, relative to a crank angle value of cylinder Z1. In the illustration of FIGS. 1C and 2which is based on the entire internal combustion engine with four cylindersthis crank angle value corresponds to 60 to 30 before top dead center of ignition (ZOT). In the following, only 660 KW to 690 KW is concerned.

[0069] FIG. 3 shows the detail X from FIG. 2, i.e. the development in the speed 101 over the crank angle KW during the diagnosis time window 112 with the limit points P1 and P2 of the diagnosis time window of cylinder Z1. The pressure pi prevails in the cylinder at point P1, and pressure p2 at point P2.

[0070] A diagnosis time point 113 in the diagnosis time window 112 is determined, for example in the middle of the diagnosis time window at 675 KW. For this time point, the temperature T* in the combustion chamber of cylinder Z1 is calculated, for example. For the determination of the diagnostic cylinder pressure p.sub.diag in dependence on the development in the speed 101, on the other hand, a time window such as the diagnosis time window 112 is required because the determination is based on a difference consideration.

[0071] FIGS. 2 to 6 explain an exemplary embodiment of methods according to the invention for determining a fresh-air mass parameter rf in a cylinder Z of the internal combustion engine 1 in driving operation with the aid of the crankshaft speed n of the crankshaft drive KT.

[0072] As shown in FIG. 6, the method carried out in the exemplary embodiment is described below: S10: Determine whether at least quasi steady-state operation SB or transient operation TB of the internal combustion engine 1 is present.

[0073] S20: If transient operation TB of the internal combustion engine is present, the cylinder Z1 which is at the end of the intake stroke or at the beginning of the compression stroke is identified.

[0074] S30: Determine the diagnosis time window 112 for the identified cylinder Z1 in the low-torque range 12 of the internal combustion engine 1.

[0075] S40: Determine the development of the speed 101 of the internal combustion engine during the specified diagnosis time window 112 with a real-time-capable sampling quality. A live engine control function continuously reads out speed values n for the crankshaft KT during driving operation (due to gas friction delay (and for the present purposes disregarded delay due to mechanical friction), an increased speed drop from one to a subsequent point in time is to be expected in a compression phase of a cylinder) and determines a development in the speed from thiscf. FIGS. 1-3.

[0076] S50: Determine the pressure characteristic p.sub.cyl,diag for the cylinder Z1 in diagnosis time window 112 in dependence on the determined development in the speed 101.

[0077] S60: Determine the simplified cylinder load parameter rf* in dependence on the determined pressure characteristic p.sub.cyl,diag for the cylinder Z1 in the diagnosis time window 112.

[0078] S70: Determine the fresh-air mass parameter rf for transient operation TB in the identified cylinder Z1 in dependence on the determined simplified cylinder load parameter rf*, in the exemplary embodiment additionally in dependence on a steady-state cylinder load parameter rf.sub.SB and/or an offset prediction rf.sub.OFFSET of the fresh-air mass parameter derived therefrom and determined in a manner known per se by means of the engine control unit for steady-state operation (cf. step S160 for steady-state operation SB). Depending on the operating state of the internal combustion enginein particular depending on the degree of transience of engine operationcontrol of the fuel quantity to be injected purely on the basis of the determined, simplified cylinder load variable may be sufficient; or the injection quantity may already be subject to feedforward control on the basis of known methods for determining the fresh air quantity in the cylinder in steady-state operation or for offset prediction on the basis of such values.

[0079] S160: A steady-state cylinder load parameter rf.sub.SB and/or an offset prediction rf.sub.OFFSET of the fresh-air mass parameter derived therefrom are determined in a manner known per se by means of the engine control unit. The step can also be carried out to support the feedforward control of the fuel injection quantity if transient operation TB is present; cf. input variables for determining the fresh-air mass parameter rf according to step S70.

[0080] S170: Determine the fresh-air mass parameter rf for steady-state operation SB in the identified cylinder Z1 in dependence on a steady-state cylinder load parameter rf.sub.SB already determined (in a manner known per se) by means of the engine control unit for steady-state operation and/or an offset prediction rf.sub.OFFSET of the fresh-air mass parameter derived therefrom. The simplified cylinder load parameter rf* is not used for steady-state operation SB.

[0081] In the exemplary embodiment, various options are provided for using the determined values of the fresh-air mass parameter rf for onboard diagnostics 204 and/or offboard diagnostics 208 and/or control tasks 206 by means of the engine control unit 2 (cf. FIG. 6).

[0082] For this purpose, the values determined are continuously stored in a non-volatile memory 202 of the engine control unit 2 during driving operation of the motor vehicle or are stored for further use. If, for example, the associated value for the fresh-air mass parameter rf is evaluated for each cylinder Z at each ignition, a new value of the fresh-air mass parameter rf is stored in the memory 202 for each ignitionin particular with a time stamp and/or output values for determining and/or specifying the diagnosed cylinder, for example Z1.

[0083] The stored values of the fresh-air mass parameter rf can be provided in real time, i.e. in particular immediately during driving operation, for example to an online diagnostic component 204 and/or an engine closed-loop control 206 of the engine control unit 2. Also, the values of the fresh-air mass parameter rf can be made available to an offboard diagnostic computer 208 at a later time, for example in the workshop.

[0084] In the following, it is explained in detailinter alia on the basis of the illustrations in FIGS. 4 and 5how the simplified cylinder load parameter rf* and, on that basis, the fresh-air mass parameter rf are determined in the exemplary embodiment.

[0085] As can be seen from FIG. 4, the following relationship applies to the composition of the gas mass in cylinder Z1 in diagnosis time window 112:

m=m.sub.tot=m.sub.air+m.sub.fuel+m.sub.residual gas(1)

[0086] The following relationship exists here between the air mass and the fuel mass:

[00001]$\begin{array}{cc}{m}_{\mathrm{fuel}}=\frac{m_{air}}{.Math.L_{\mathrm{st}}}& \left(2\right)\end{array}$

TABLE-US-00001 Formula symbol Meaning measured combustion air ratio (<1 = rich, 1 = stoichiometric, >1 = lean) L.sub.st fuel-dependent chemical constant, so- called stoichiometric fuel-air ratio, typically between 14-16

[0087] Equation (2) in (1) gives

[00002]$\begin{array}{cc}{m}_{tot}=m_{\mathrm{air}}.Math.\left(1+\frac{1}{.Math.L_{st}}\right)+m_{\mathrm{residual}\mathrm{gas}}& \left(3\right)\end{array}$

[0088] In the exemplary embodiment, a substitution of the residual gas mass takes place via typical engine control variables:

m.sub.residual gas=xrg.Math.m.sub.tot(4)

[0089] The residual gas mass can be interpreted as fraction xrg of the total mass.

[0090] In order to perform a substitution of the absolute air mass in equation 3, the following relationship is introduced based on typical engine control variables:

[00003]$\begin{array}{cc}{m}_{\mathrm{air}}={\mathrm{rf}}_{SB}.Math.\frac{p_0.Math.V_{\max }}{R.Math.T_0}& \left(5\right)\end{array}$

TABLE-US-00002 Formula symbol Meaning rf.sub.SB Steady-state fresh-air mass parameter, relative air charge of the cylinder in % p.sub.0 atmospheric pressure under standard conditions (1013 hPa) V.sub.max maximum cylinder volume at bottom dead center of the crankshaft R ideal gas constant T.sub.0 ambient temperature under standard conditions (293 K)

[0091] The current air mass in the cylinder is determined in advance in the engine control unit as the steady-state fresh-air mass parameter rf.sub.SB for the purpose of correct fuel addition.

[0092] The function known per se and already present in the engine control unit for this purpose is the so-called load detection for steady-state engine operating states. It estimates a relative filling in percent.

[0093] The aim of the exemplary method described here is to improve the estimation of the reference variable rf. (The filling rf is defined as 100% if the max. cylinder volume were completely filled with air under standard conditions, cf. ideal gas equation):

[00004]$\begin{array}{cc}{m}_{\mathrm{air}}=\mathrm{rf}.Math.\frac{p_0.Math.V_{\max }}{R.Math.T_0}& \left(5.5\right)\end{array}$

[0094] The total cylinder mass in turn results from the current thermodynamic ratios of cylinder pressure p*, cylinder volume V* and temperature T* in the cylinder, since the cylinder is not only filled with air and the components of fuel and residual gas lead to an increase in pressure:

TABLE-US-00003 Formula symbol Meaning p* Cylinder pressure in diagnosis time window V* Cylinder volume at time of diagnosis R ideal gas constant T* Temperature T* in the cylinder at the time of diagnosis

[00005]$\begin{array}{cc}{m}_{\mathrm{tot}}=\frac{p^{*}.Math.V^{*}}{R.Math.T^{*}}& \left(6\right)\end{array}$

[0095] Insertion of (6), (5.5) and (4) into (3) including rearrangement and truncation leads to this relationship:

[00006]$\begin{array}{cc}{p}^{*}.Math.V^{*}.Math.\frac{1-\mathrm{xrg}}{T^{*}}=\mathrm{rf}.Math.\frac{p_0.Math.V_{\max }}{T_0}.Math.\left(1+\frac{1}{.Math.L_{st}}\right)& \left(7\right)\end{array}$

[0096] Based on FIGS. 5A-E, it is explained below howstarting from equation (7) a simplified relationship is established which manages with a smaller number of variable quantities and thus also with significantly lower computing power in the engine control unit, so that a simplified cylinder load parameter rf* can be determined in real time. In this exemplary embodiment, real-time capability means that the simplified cylinder load parameter rf* can be used to determine a fuel injection quantity for the next cycle on the basis of the values determined for one cycle.

[0097] FIGS. 5A-F show a graphical derivation of simplifying assumptions for the interrelationships of the cylinder content parameters and the state variables.

[0098] Starting from the complete relationship shown in FIG. 5A, a further simplification is introduced with each further FIG. 5B, 5C, 5D and 5E, so that finally in FIG. 5F a simplified relationship is shown which nevertheless still permits a statement accuracy sufficient for the purposes of the invention.

[0099] The simplifications for equation (7) are aimed at parameterizing the residual gas fraction xrg and the cylinder temperature T*.

[0100] FIG. 5A shows the full relationship of the variables. The line thickness represents the correlation strength. Each line is considered in first approximation as an approximation of a proportionality relation, in order to simplify again the existing equation set later. Dashed lines indicate inverse proportionality (and are marked indirect accordingly).

[0101] The cylinder Z1 is filled with a fresh-air mass m.sub.air, which is represented by the fresh-air mass parameter rf. In addition, the cylinder is filled with the fuel mass mruei and a residual gas mass m.sub.residualgas, which is represented by the residual gas fraction xrg.

[0102] Indirectly or directly, all three parameters of the cylinder content act on at least one of the two relevant state variables of the mixture in the cylinder Z1, namely p* and V*.

[0103] The residual gas fraction xrg has a medium influence on the total mass m.sub.tot in the cylinder; likewise on the temperature T*. The residual gas fraction xrg also has a small influence on the pressure p* in the cylinder. Both are known from experimental observations and can be regarded as generalizable.

[0104] The fresh-air mass parameter rf has a major influence in each case on the total mass m.sub.tot in the cylinder and thus also on the fuel mass m.sub.fuel.

[0105] The total mass m.sub.total in the cylinder in turn has a large influence on the cylinder pressure p* via the ideal gas equation.

[0106] The cylinder pressure p* in turn has a large influence on the temperature T* star in the cylinder.

[0107] FIG. 5B shows a change in the use of the residual gas fraction xrg to an inversely proportional consideration to allow for a later simplification step in which an indirect influence of the residual gas fraction on the cylinder pressure is introduced (cf. Figure D).

[0108] In FIG. 5C, a removal of weak connections and then solitary adjacent elements occurs.

[0109] In FIG. 5D, the intermediate quantity m.sub.tot is substituted as shown.

[0110] In FIG. 5E, a weak connection resulting from the substitution step is removed.

[0111] In FIG. 5F, only a conversion to the temperature T* as the target variable is shown. Since each line was considered as an approximation of a proportionality relation, two borrowable substitution equations result from the diagram shown. The first substitution equation is:

T*=C2.Math.p*(8)

[0112] With the further relationship

(T*).sup.C1.Math.(1xrg).sup.1=C0|C1>1

this gives

(C2.Math.p*).sup.C1.Math.(1xrg)=C0

and by combining the constants

p*.sup.C1.Math.(1xrg)=C3

or converted to the second substitution equation

1xrg=C3.Math.p*.sup.C1(9)

[0113] Equations (8) and (9) are now used for the corresponding variables of equation (7) and, in addition, an amalgamation of the constants is provided:

[00007]$\begin{array}{cc}C4.Math.p^{*C5}.Math.V^{*}=\mathrm{rf}.Math.\frac{p_0.Math.V_{\max }}{T_0}.Math.\left(1+\frac{1}{.Math.L_{st}}\right)& \left(10\right)\end{array}$

[0114] The determination of the constants C4, C5, etc. in the model equations was carried out in the exemplary embodiment on the development engine with the aid of the following procedure: a complete characteristic map (speed/load) is measured; evaluation of cylinder indexing p* and calculation of xrg and T* via corresponding gas exchange analyses; then, accordingly, a calculation of the respective characteristic values is carried out from the results and plotting over mean engine speed (characteristic curve).

[0115] The combining of the constant C4 with the fixed values p.sub.0, T.sub.0 and V.sub.max gives in the following:

[00008]$\begin{array}{cc}C6.Math.p^{*C5}.Math.V^{*}=\mathrm{rf}.Math.\left(1+\frac{1}{.Math.L_{st}}\right)& \left(11\right)\end{array}$

[0116] Lastly, it is now possible to convert to rf and thus derive the determination rule for the relative load, in this case initially the simplified fresh-air mass parameter rf*:

[00009]$\begin{array}{cc}{\mathrm{rf}}^{*}=C6.Math.\frac{p^{*\mathrm{C5}}+C7}{1+\frac{1}{.Math.L_{st}}}.Math.V^{*}& \left(12\right)\end{array}$

[0117] The constant C7 was introduced subsequently in the application of equation (12) to make the model as adaptable as possible. (The constant C7 can also be assumed to be C7=0 in the initial application and can later take on other values accordingly for improved model accuracy).

[0118] Below is a table for determining the open parameters for rf estimation:

TABLE-US-00004 Value Unit Description C6 [%/Nm] Scaling factor: working term to load C5 [] Exponential scaler: pressure to load C7 [bar] Offset: pressure to load (Default = 0)

[0119] The values for , L.sub.St and V*, in each case for a time defined by the crankshaft position of the diagnosis time point 113, can be taken from known engine control units, including that of the exemplary embodiment.

[0120] A diagnostic cylinder pressure value p.sub.cyl,diag for the diagnosis time window 112 is determined as the value for p*.

[0121] How this is possible can be taken from the following description for equations (13)(28), wherein, from the determined diagnosis time window (see explanations for FIG. 2) .sub.1 corresponds to the crank angle KW=660.sup.0 of P1 and p2 corresponds to the crank angle KW=690 of P2, and accordingly in the shown exemplary embodiment p.sub.cyl,diag =p.sub.cyl1, diag, 660-690.

[0122] The determination is based on a pressure balancing of the diagnosed cylinder on the basis of the measured speed curve:

[00010]$\frac{d}{\mathrm{dt}}\left(\frac{1}{2}.Math.J_0.Math.{}^2\right)=\left(M_{\tan }-M_R-M_L\right).Math.$

TABLE-US-00005 Formula symbol Meaning J.sub.0, J General/proportional mass moment of inertia Angular position of crankshaft Angular velocity M.sub.tan Moment due to gas force in cylinder and oscillating mass force M.sub.R Moment due to friction losses M.sub.L Moment due to load reduction M.sub.M Proportional moment due to rotational mass inertia n.sub.mot Currently applied motor speed

[0123] By differentiation, substitution and introduction of a mass moment (division of inertia components), the following equation is obtained:

[00011]$J.Math.\stackrel{.}{}=\underset{i}{.Math.}{M}_i=M_{\tan }-M_R-M_L-M_M$

[0124] If the equation is divided sensibly into a constant component and an alternating component, the following sub-equations are obtained:

Constant component: M.sub.tan=M.sub.RM.sub.L

[0125] The balancing of the constant component assumes a steady-state operating point. The mean provided torque keeps the mean speed constant because it corresponds to the torque demands from load and friction.

Alternating component: J.Math.{dot over ()}={tilde over (M)}.sub.tan{tilde over (M)}.sub.R{tilde over (M)}.sub.M(13)

[0126] A conversion from time-based derivation to crank-angle-based differencing is performed using the relationship

[00012]$\begin{array}{cc}=\frac{d}{\mathrm{dt}}=.Math.\frac{n_{\mathrm{mot}}}{30}\mathrm{per}& \left(14\right)\end{array}$$\stackrel{}{}{\left(\frac{}{30}\right)}^2.Math.\mathrm{nmot}.Math.\frac{\mathrm{nmot}}{}$

[0127] The decisive quantities from equation (13) are further detailed for the evaluation. The relationship for the resulting moment from the inner-cylindrical gas force results in:

[00013]$\begin{array}{cc}=\left[A_K.Math.\left[P_{\mathrm{cyl}}\left(\right)-p_0\right]-m_{\mathrm{osc}}.Math.\stackrel{.Math.}{s}\left(\right)\right].Math.r_K.Math.\frac{\sin \left(+\right)}{\cos }& \left(15\right)\end{array}$

TABLE-US-00006 Formula symbol Meaning A.sub.K Piston top surface = const. r.sub.K Effective radius of the crankshaft corresponds to half stroke = const. l.sub.Pl Connecting rod length = const. m.sub.osc Oscillatory mass part corresponds to piston assembly and proportional connecting rod mass = const. p.sub.cyl Pressure prevailing in cylinder p.sub.0 Reference pressure, crankcase pressure B() Connecting rod pivot angle in dependence on crank angle position {umlaut over (s)}() Piston acceleration in dependence on piston position

[0128] A further detailing of the variable factors from equation (15) gives:

{umlaut over (s)}(,{dot over ()},{umlaut over ()})=r.sub.K.Math.sin+r.sub.K.Math.{dot over ()}.sup.2.Math.cosr.sub.K/.sub.2.sub.Pl.Math.sin(2.Math.)+r.sub.K.Math.{dot over ()}.sup.2.Math..sub.Pl.Math.cos(2.Math.)

[0129] Assuming a constant mean speed nmot, the relationship for the piston acceleration becomes simpler:

{umlaut over (S)}.sub.red(,{dot over ()})=r.sub.K.Math.{dot over ()}.sup.2.Math.(cos +.sub.pl.Math.cos(2))(16)

[0130] The assumption leads to an error that can be disregarded. The influence of the angular acceleration results in a negligible deviation over the entire characteristic map.

()=arcsin(.sub.pl.Math.sin)(16.5)

[0131] Push rod ratio

Pl=r.sub.K/l.sub.Pl(17)

P.sub.cyl=P.sub.cyl(18)

[0132] Reference to ambient pressure

P.sub.0=P.sub.amb(19)

or as also used in the following the reference to crankcase pressure

P.sub.0=P.sub.Crkc=P.sub.ambDPS(20)

wherein DPS stands for the negative pressure (pressure difference) in the intake manifold.

[0133] The frictional torque from equation (13) can be represented in different ways. Either a model can be introduced which reflects measured data for a specific operating point of the diagnosis. A target-oriented approach here would be a functional linking of the term with the speed, the load and the oil temperature.

[0134] In the following, however, it is assumed that the diagnosis is carried out at fixed, steady-state load points. This means that the frictional torque for this load point can be assumed to be invariable.

{tilde over (M)}.sub.R=const.(21)

[0135] The same approach is also used for the proportional moment due to rotational inertia and the mass moment of inertia.

.sub.M=const.(22)

J=const.(23)

[0136] A suitable choice of diagnostic constants at the steady-state operating point allows easy application of the parameters in retrospect.

[0137] Solving equation (13) according to the gas moment gives:

=J.Math.{dot over ()}++

[0138] After inserting the relationships from equations (21) to (23), the following simplification with the application constant K_RM can be concluded:

=J.Math.+K.sub.RM(24)

Application of the Diagnosis:

[0139] FIG. 3 shows the detail X from FIG. 2, i.e. the development in the speed 101 over the crank angle KW during the diagnosis time window 112 with the measuring points P1 and P2 in the compression of cylinder Z1. Pressure pi prevails in the cylinder at point P1, and pressure p2 at point P2.

[0140] The gradient of the angular velocity from equation (14) is expanded. The speed to be determined must be averaged here, and constants are marked again.

[00014]$\begin{array}{cc}\stackrel{}{}{\left(\frac{}{30}\right)}^2.Math.\stackrel{\_}{n_{\mathrm{mot}}}.Math.\frac{n_{\mathrm{mot}}}{}& \left(25\right)\end{array}$$\stackrel{}{}{\left(\frac{}{30}\right)}^2.Math.\frac{n_{{\mathrm{mot}}_2}+n_{{\mathrm{mot}}_1}}{2}.Math.\frac{n_{{\mathrm{mot}}_2}-n_{{\mathrm{mot}}_1}}{{}_2-{}_1}$$\stackrel{}{}\frac{1}{2}.Math.{\left(\frac{}{30}\right)}^2.Math.\frac{n_{{\mathrm{mot}}_2^2}-n_{{\mathrm{mot}}_1^2}}{{}_2-{}_1}$$\stackrel{.}{}{K}_{}.Math.\frac{n_{{\mathrm{mot}}^2}-n_{{\mathrm{mot}}_1^2}}{{}_2-{}_1}$

[0141] The term for the tangential moment from equation (15) is expanded in the following by the relationships from equations (16) to (20), and constants are marked.

[00015]$\begin{array}{cc}{\tilde{M}}_{\tan }=\left[\frac{\left(p_1+p_2-2.Math.p_{\mathrm{amb}}+2.Math.\mathrm{DPS}\right)}{2}.Math.A_K-m_{\mathrm{osc}}.Math.\stackrel{.Math.}{s}\left(\right)\right].Math.r_K.Math.\frac{\sin \left(+\right)}{\cos }& \left(25.5\right)\end{array}$${\tilde{M}}_{\tan }=\left[\frac{\left(p_1+p_2-2.Math.p_{\mathrm{amb}}+2.Math.\mathrm{DPS}\right)}{2}.Math.A_K-m_{\mathrm{osc}}.Math.\stackrel{.Math.}{s}\left(\right)\right].Math.K_K$

with a kinematic constant for the steady-state point in which the diagnosis takes place

[00016]$\begin{array}{cc}{K}_K=r_K.Math.\frac{\sin \left(+\right)}{\cos }& \left(26\right)\end{array}$

[0142] After inserting equations (26) and (25) into equation (24), resolving according to the cylinder pressures, and amalgamating all constants, the following is given:

[00017]$\begin{array}{cc}\left[\frac{\left(p_1+p_2-2.Math.p_{\mathrm{amb}}+2.Math.\mathrm{DPS}\right)}{2}.Math.A_K-m_{\mathrm{osc}}.Math.\stackrel{.Math.}{s}\left(\right)\right].Math.K_K=J.Math.K_{}.Math.\frac{n_{{\mathrm{mot}}_2^2}-n_{{\mathrm{mot}}_1^2}}{{}_2-{}_1}+{}_{\mathrm{RM}}& \left(27\right)\end{array}$$\frac{p_1+p_2}{2}=\frac{J.Math.K_{}}{K_K.Math.A_K}.Math.\frac{n_{{\mathrm{mot}}_2^2}-n_{{\mathrm{mot}}_1^2}}{{}_2-{}_1}+\frac{K_{\mathrm{RM}}}{K_K.Math.A_K}+\frac{m_{\mathrm{osc}}.Math.\stackrel{.Math.}{s}\left(\right)}{A_K}+p_{\mathrm{amb}}-\mathrm{DPS}$$\frac{p_1+p_2}{2}=K_1.Math.\frac{n_{{\mathrm{mot}}_2^2}-n_{{\mathrm{mot}}_1^2}}{{}_2-{}_1}+K_2+\frac{m_{\mathrm{osc}}.Math.\stackrel{.Math.}{s}\left(\right)}{A_K}+p_{\mathrm{amb}}-\mathrm{DPS}$

[0143] All pressure variables and speeds in equation (27) can be measured at the times P1 and P2 for the conditions of the constants shown. A suitable indexing measurement technique, known per se, resolves the necessary physical quantities based on the crank angle or at least averaged over several operating cycles. In addition or as an alternative to the indexing measurement technique, data from a suitable operating model, for example the motor control system, can be used. The kinematic constant K.sub.K can be tabulated and used in dependence on the piston position.

[0144] The influence of the speed n.sub.mot related to the oscillatory masses can, for example, be calculated in real time or stored on the control unit in the form of a lookup table of a suitably stored operating model with respect to speed and load.

[0145] The reduced piston acceleration (cf. in particular equation (16)) can be formulated for the two discrete points:

[00018]$\begin{array}{cc}{\stackrel{.Math.}{s}}_{\mathrm{red}}\left(,\stackrel{}{}\right)=r_K.Math.{\left[\frac{\frac{}{30}{n}_{{\mathrm{mot}}_1}+n_{{\mathrm{mot}}_2})}{2}\right]}^2.Math.\left[\cos \left(\frac{{}_1+{}_2}{2}\right)+\cos \left({}_1+{}_2\right)\right]& \left(27.5\right)\end{array}$

[0146] The constants K.sub.1 and K.sub.2 can be determined on the basis of reference measurements (motor function and load change OK, respectively).

[0147] After determining the application constants K.sub.1 and K.sub.2, equation (27) can be used to determine the diagnostic cylinder pressure from the speed change in the compression:

[00019]$\begin{array}{cc}{\stackrel{}{p}}_{\mathrm{cyl},\mathrm{diag}}=K_1.Math.\frac{n_{{\mathrm{mot}}_2^2}-n_{{\mathrm{mot}}_1^2}}{{}_2-{}_1}+K_2+\frac{m_{\mathrm{osc}}.Math.{\stackrel{.Math.}{s}}_{\mathrm{red}}\left(,n_{\mathrm{mot}}\right)}{A_K}+p_{\mathrm{amb}}-\mathrm{DPS}& \left(28\right)\end{array}$

[0148] The diagnostic cylinder pressure p.sub.cyl,diag is an indicator of the pressure curve during the compression stroke of the cylinder.

[0149] In this way, the diagnostic cylinder pressure p.sub.cyl,diag =p.sub.cyl, diag, 660-690 can be determined for the diagnosis time window 112 of the diagnosed cylinder Z at the time interval t12=t[P1; P2] in driving operation.

[0150] This calculation of the diagnostic cylinder pressure p.sub.cyl, diag, 660-690 in the calculated operating cycle for the calculated cylinder is used to estimate the simplified cylinder load parameter rf* in the next operating cycle according to equation (12).

[0151] The fresh-air mass parameter rf can then also be determined from this, if necessary, in dependence on the steady-state cylinder load parameter rf.sub.SB determined for steady-state operation and/or an offset prediction rfoFFSET derived from this. Weightings with which the simplified fresh-air mass parameter rf*, the steady-state fresh-air mass parameter rf.sub.SB and/or the offset prediction rf.sub.OFFSET are included in the calculation of rf for transient operating states TB are in themselves dependent on the degree of transience and/or other expert considerations considered on their own.

[0152] Feedforward control of the fuel injection quantity into the cylinder Z1 then takes place in the exemplary embodiment for an operating cycle on the basis of the value of the fresh-air mass parameter rf determined for the previous operating cycle.

LIST OF REFERENCE SIGNS

[0153] 1 Internal combustion engine [0154] 2 Control means [0155] 4 Computing unit [0156] 6 Detection unit for the speed of the crankshaft [0157] 7 Cylinder pressure detection unit [0158] 9 Intake system [0159] 10 Torque curve of the internal combustion engine over an engine cycle [0160] 12 Low-torque ranges [0161] 14 Predetermined limit for relevant torque contribution [0162] 16 Cylinder temperature detection unit [0163] 18 Lambda sensor [0164] 150 Graph showing development in the speed [0165] 101 Speed curve [0166] 112 Diagnosis time window [0167] 113 Diagnosis time point [0168] 200 Engine control unit [0169] 202 Memory [0170] 204 Diagnostic component of an engine control unit [0171] 206 Control component of an engine control unit [0172] 208 Offboard diagnostic computer [0173] KT Crank drive [0174] KW Crank angle [0175] L.sub.St Stoichiometric fuel-air ratio, fuel-specific [0176] m.sub.fuel Fuel mass in the cylinder [0177] m.sub.air Air mass in the cylinder [0178] m.sub.residual gas Residual gas mass in the cylinder [0179] m.sub.tot Gas mass in the cylinder [0180] M Torque of a cylinder in FIG. 1 [0181] n Speed [0182] p* Cylinder pressure at time of diagnosis [0183] P.sub.cyl,diag Pressure index, here diagnostic cylinder pressure [0184] P Measurement time points at the beginning and end of the diagnosis time window [0185] p.sub.0 Atmospheric pressure under standard conditions (1013hPa) [0186] R ideal gas constant [0187] rf Fresh-air mass parameter; relative air charge of the cylinder in %. [0188] rf* Simplified cylinder load parameter [0189] rf.sub.SB Steady-state cylinder load parameter [0190] rf.sub.Offset Offset prediction [0191] SB Steady-state operation [0192] Time interval in the diagnosis time window [0193] T* Temperature of the gas mixture in the cylinder at time of diagnosis [0194] T.sub.0 Ambient temperature under standard conditions (293K) [0195] TB Transient operation [0196] V* Cylinder volume at time of diagnosis [0197] V.sub.max maximum cylinder volume at bottom dead center of crankshaft [0198] xrg Residual gas fraction [0199] Z Cylinder [0200] ZZP Ignition timing point of a cylinder [0201] Combustion air ratio