EP1242738B1 - Regulation of true running for diesel engines - Google Patents

Description

The invention relates to a method for concentricity control, such as that from DE 195 48 604 C1 emerges as known. The known method serves Differences in the torque contributions of individual cylinders of an internal combustion engine based on to determine the crankshaft speed curve. It builds on the knowledge that the rotary motion of the crankshaft under the action of gas and mass forces runs irregularly. To the speed or torque component of a cylinder determine, individual cylinders are specifically switched off during engine operation. By comparison with the speed curve of the operated without cylinder deactivation Engine, the torque percentage of each individual cylinder can be Display the total engine torque in isolation based on the speed signal. The of Production tolerances resulting from injection quantity variations are recognized and should be can be compensated by using the same mean pressures in all cylinders Injection quantity variation can be produced.

A similar process is described in DE 41 22 139 C2. Here too it will assumed that rotational nonuniformities occur due to the fact that of tolerances in the injectors into the individual cylinders of the Internal combustion engine different amounts of fuel are injected. Approach is that the torque or the rotational acceleration is directly proportional to the injected Amount of fuel. To avoid the speed irregularities, the portion of every combustion process at the spin. The measured values are compared with each other by forming averages and in this way Discrepancies detected. The fuel injection quantities of the individual cylinders are finally changed so that the deviations disappear. The sum of the Changes in the amount of fuel injected into the individual cylinders are selected so that it gives a total of zero.

In an internal combustion engine according to WO 97/23716, the fuel supply can be one Cylinder are switched off, which then works as a compressor, for example. To in In this mode of operation to avoid vibrations, the fuel supply is provided to change the remaining, normally working cylinders in a suitable manner. It should be possible to determine through experimentation and calculation in which way the Torque of the cylinder is to be distributed in order to optimally suppress the To achieve vibrations. For certain operating cases, this way determined data available, according to which the internal combustion engine is controlled. The Injection quantities are obviously distributed among the individual cylinders so that the Vibrations of the 0.5th to 3rd orders are suppressed, since only they in the Practice are responsible for noticeable vibrations. However, the Obviously, vibrations of the different orders are not always equally suppress. The appropriate fuel distribution is apparently related to the Size of the vector responsible for the vibrations.

WO 98/07971 also describes a method for cylinder-selective control of a self-igniting internal combustion engine as known. One serves Measuring device for detecting the crankshaft rotation angle and for determining the current crankshaft speed. Determined from the crankshaft speed Control unit suitable parameters that in different operating areas of the Internal combustion engine a cylinder-selective equality or a defined Enable inequality of mean pressures, with the effect of Component differences in the fuel supply and the combustion system on the Combustion process can be minimized.

In the dissertation by Jochen Tonndorf: "Influence of misfiring on the Torsional vibration behavior of drive systems with piston engines ", approved by the Faculty of Mechanical Engineering of the Rheinisch-Westfälische Technische Hochschule Aachen the torsional vibration behavior of motors is examined. It is stated that there are operating conditions that differ significantly from normal operation. So lead Tolerance-related manufacturing differences for cylinders and injection devices, but also differences due to wear and tear in the course of the operating time compared to normal operation. This can allegedly result in deviations in performance individual cylinders of about +/- 10%, which is the origin of a Torsional vibration excitation force causes. In particular, multi-cylinder Motors add up the deviations of the individual cylinders so unfavorably that the Impact is the same as if a cylinder has completely failed. Furthermore, interruptions in operation occur due to faults in the injection system. Damaged one or Exhaust valves can result in loss of compression. Switching off too of cylinders represents an operating case, the torsional vibration stress changed. The effect of operating conditions that deviate from normal operation the excitation behavior of the motor is shown by a vector representation of the excitation forces clarified. It is further stated that only the excitatory ones in dropout operation Forces of the 0.5th, 1st and 1.5th order are of interest. The exciting one Alternating torque is calculated from the vector sum according to the phase position the harmonic. However, the author concludes that engine interventions, e.g. are practically not feasible by changing the ignition pressure.

The invention has for its object a concentricity control especially for to represent high-cylinder internal combustion engines.

This object is achieved by the features listed in claim 1. While in internal combustion engines with a few cylinders, the individual cylinders declining speed components clearly in the speed curve of a work cycle can be seen, this is not the case with high-cylinder internal combustion engines. Rather, the speed components overlap in such a way that when the Speed curve it is no longer possible to draw any conclusions about the cylinder causing it, which requires new evaluation methods. Nonetheless, the inventive method is also apply to low-cylinder internal combustion engines, if there too There are restrictions due to the small number of cylinders. For the concentricity control the low-frequency vibrations are considered. For this, the Pulse response spectrum of each cylinder determined by calculation or measurement. to Determination of the momentum component of a cylinder in the speed of rotation by measurement the cylinders are switched off one after the other and the speed above Crank angle recorded. In addition, the speed curve of the healthy is intact Motors, that is, when all cylinders work normally. It can be a brand new engine in normal operation, due to tolerances has slight differences in the speed components of each cylinder, or by one ideal engine whose cylinder, for example, by using the invention Procedure are equal in terms of their shares in the speed acceleration.

In this sense, ideal means that before the reference values are recorded, for example by varying the injection quantities of individual cylinders, a setting is made in which the fluctuations in the speed contributions of the cylinders are minimized. This setting is retained in normal operation. By forming the difference between the curve shape of the healthy engine and the curve shapes for individually deactivated cylinders, new curves are then generated which reflect the influence of each cylinder on the overall speed curve. These response curves are subjected to a Fourier decomposition. However, only low-frequency harmonic vibrations, expediently the 0.5th to 3rd order, and the associated spectral impulse responses are considered I of the speed curve of a working cycle of each cylinder. In normal engine operation, the speed curve of the crankshaft is now continuously recorded over the angle and the spectrum of the speed curve is analyzed in an analogous manner by Fourier decomposition of the curve curve obtained R of a working game. To represent the spectral speed curve, only the Fourier coefficients of the low-frequency vibrations are used, namely preferably the harmonics of the 0.5th to 3rd order, which are processed to form a line matrix. The spectral impulse responses I and the resultant from Fourier coefficients of the speed curve R can be represented for each harmonic as a vector pointer over the crank angle. If the resultant is zero, no correction of the injection quantities is necessary. However, if there is a resultant, this means that there is a reduced injection in a cylinder, and the resultant must be made zero by correcting the injection quantities of the individual injectors. The division of the total injection quantity required for the given load case takes place in such a way that the components of the resultant in the direction of the impulse response pointers with the impulse responses I be multiplied. The result is correction factors for the injection quantities. Cylinders pointing towards the resultant R are corrected more with a positive or negative sign than orthogonal ones. The mathematical operation that can accomplish the corresponding achievement is the formation of the dot product or the vectorial product from the resultant R and the spectral impulse responses I. The necessary data are available in matrix form for this. The matrix multiplication of impulse responses I with the vector of the spectral speed curve R results in values different from zero and leads to a correction of the injection quantities if there is a runout deviation in normal operation. The correction values, which are standardized, are fed to a controller and the injection quantities ΔQ are determined, which can be positive or negative and accordingly correct the injection quantities determined by the engine controller for each injector of a cylinder.

Figur 1: Einen Drehzahlregelkreis mit den für die Drehschwingungsanalyse notwendigen Elementen in schematischer Darstellung;

Figur 2: Den Drehzahlverlauf der Kurbelwelle über dem Winkel für ein Arbeitsspiel des Motors;

Figur 3: Eine spektrale Darstellung der Impulsantwort I eines Zylinders;

Figur 4: Eine Zeigerdarstellung der Drehzahlanteile der Zylinder an der 0,5-ten Ordnung für einen Sechszylinder-Motor und zwar für einen gesunden Motor (Figur 4a), einen Motor mit fehlendem Injektor (Figur 4b) und für einen Motor mit korrigierter Einspritzmenge (4c).

The invention is illustrated by means of the drawings with FIGS. 1 to 4. The figures show:

Figure 1: A speed control loop with the elements necessary for torsional vibration analysis in a schematic representation;

Figure 2: The speed curve of the crankshaft over the angle for a working cycle of the engine;

Figure 3: A spectral representation of the impulse response I a cylinder;

FIG. 4: A pointer representation of the speed components of the cylinders in the 0.5th order for a six-cylinder engine, specifically for a healthy engine (FIG. 4a), an engine with no injector (FIG. 4b) and for an engine with corrected injection quantity ( 4c).

In Figure 1, a speed control loop is shown, as it is known for example from DE 195 15 481 A1. Reference numeral 1 denotes a diesel engine, the crankshaft, not shown, is connected to a measuring wheel 2. With the measuring wheel 2 and a sensor 3, the speed curve of the crankshaft can be recorded over the angle. With a filter 4 and a filter 5, faults are masked out and the curve shape is averaged by comparing the recorded curve shapes over several work cycles. For concentricity control, the speed curve of the crankshaft is continuously recorded over the angle in normal engine operation. The speed signal of a work cycle is shown by way of example in FIG. 2. The radius marked with r corresponds to the current speed at the angle . The speed curve shows a deformation that occurs when a cylinder fails. By Fourier decomposition of the speed curve, the spectral speed curve is obtained with the resulting vectors R ₁ to R _n , where the indices correspond to the considered harmonics. The corresponding operation is carried out in the symbolically represented function block 7. The vectors obtained by Fourier decomposition R are the Fourier coefficients. Preferably only the harmonic vibrations of the 0.5th to 3rd order are considered. With ideal concentricity, no resulting parts of the corresponding harmonics occur or are at least negligible. In reality, however, there is a small resulting vector R , because the harmonic components are not evenly distributed over the circumference. This case is shown for an engine with six cylinders as an example for the harmonic of the 0.5th order in Figure 4a. Each cylinder makes approximately the same contribution to the rotational acceleration as the vector pointer I 1 to I 6 clarify. In this case, the injection quantities determined on the basis of the predetermined target and actual speeds in the speed controller 9 and by the injection software 10 are not corrected by the injectors 11 assigned to each cylinder.

However, the injection quantity must be corrected individually for each cylinder if, as shown in FIG. 4b, a resultant due to the low-frequency vibration components R is not zero. In the corresponding case, it is assumed that a cylinder has failed and a harmonic of the 0.5th order occurs, which has the phase position shown with respect to the cylinders.

Bildung des Skalarprodukts: R ^T* I = K oder:

R ^T = Spektrum des Drehzahlverlaufs eines Arbeitsspiels (Transponierte)

I = Spektrale Impulsantworten

K = Korrekturfaktoren für die Einspritzmenge

In order to be able to calculate suitable correction factors for the injection quantities of the injectors in order to produce the concentricity, the pulse share of the speed of each cylinder must be known. The corresponding speed-dependent data are kept ready in function block 8. To determine the momentum component of a cylinder in the speed of rotation, the cylinders are switched off one after the other in a measuring run and the speed is recorded over the crank angle. By comparing the speed curve of the healthy engine, the difference between the two curve curves gives new curve curves that give the impulse responses I of the engine on the cylinder shutdown. The impulse responses I are subjected to a Fourier transformation, taking the spectral impulse responses I receives. Only the parts due to the low-frequency harmonic vibrations of the 0.5th to 3rd order are considered. The spectral impulse response I = ( I _0.5 , I _1.0 , I _1.5 , I _2.0 , I _2.5 , I _3,0 ) of a cylinder is shown in Figure 3. The vector pointers indicate the amount and phase of the corresponding harmonic. The impulse responses I are stored in matrix form for mathematical processing. By forming the scalar in-product of the resulting vectors R with the impulse responses I correction factors for the injection quantities of the individual injectors are generated. This takes place in the multiplication point 13. The scalar vector product has the effect that only the components of the resultants lying in the direction of the impulse response vectors R make a contribution to the correction factors, that is to say that collinear vectors are strongly corrected and orthogonal vectors are not corrected at all. The correction values are entered in the form of vector arrows for the individual injectors in FIG. 4c. The correction factors are converted by multiplication by a constant factor into injection quantities ΔQ for each injector, which can be positive or negative, and accordingly the injection quantity Q determined by the engine controller for each injector of a cylinder is corrected positively or negatively in a summation point 12.
The calculation is based on the following equations:

Formation of the dot product: R ^T * I = K or:

R ^T = spectrum of the speed curve of a work cycle (transposed)

I = Spectral impulse responses

K = correction factors for the injection quantity

By multiplying the scalar size K by the unit vector e _I the impulse response K receive: K = K * e l

Claims (9)

1. Method for regulating the smooth running of the crankshaft of an internal combustion engine, the contributions of the individual cylinders of the internal combustion engine to the rotational acceleration being determined with the aid of the crankshaft speed characteristic, and the injection quantities of the injectors assigned to the cylinders being varied for the purpose of setting defined speed contributions to the speed characteristic, characterized in that a pulse response spectrum I of a working cycle is formed for each cylinder, at least for the harmonic of the 0.5th order, on the basis of calculated or measured crankshaft speed characteristic curves, in that in normal operation in each case the crankshaft speed characteristic is recorded over the angle of a working cycle, and the Fourier coefficients are determined by Fourier transformation as resultants R at least of the harmonic of the 0.5th order, and in that correction factors are subsequently obtained for the injection quantities of the individual cylinders by multiplying the components of the resultants R lying in the direction of the pulse response vectors by the pulse responses I and combining them by addition.
2. Method for regulating smooth running according to Claim 1, characterized in that the pulse response spectrum I is obtained from the difference between the speed curve of the healthy engine and the speed curve of the engine with in each case one cylinder cut off for each cylinder by Fourier transformation of the differential speed curve.
3. Method according to Claim 1 or 2, characterized in that the scalar product is formed from the pulse responses I and the Fourier coefficients R , the terms of which product represent in magnitude and direction, after multiplication by the unit vector, the correction factors for the injection quantities of each cylinder.
4. Method according to Claim 1, 2 or 3, characterized in that the low-frequency components of a plurality of harmonics are determined by Fourier transformation from the courses of the curves, and correction factors for the injection quantities of each cylinder are represented therefrom.
5. Method according to Claim 4, characterized in that the harmonics of the 0.5th to 3th order are considered.
6. Method according to Claim 4, characterized in that the Fourier coefficients of the 0.5th and 1th order are used.
7. Method according to Claim 5, characterized in that the harmonics of the 1.5th order are additionally taken into account.
8. Method according to one of Claims 1 to 7,
   characterized in that the coefficients of the Fourier transformations are stored and processed in the form of matrices in an onboard computer.
9. Method according to one of Claims 1 to 8, characterized in that the setting of the injection quantities of the individual cylinders of the healthy engine is corrected until the contributions of the cylinders, at least as regards low-frequency harmonics, are largely equal to the rotational acceleration, and in that the contributions of the individual cylinders to the speed characteristic are determined in relation to this speed characteristic.

Images