EP1818528A2 - Method for estimating the amount of injected fuel - Google Patents

Abstract

A process estimates the amount of fuel injected in a single operation to single cylinder in a fuel-injected multi-cylinder automotive piston engine. The segmented time-lapse in which the single injection takes place, is numerically evaluated by generating a second time-lapse derivation. The second segmented time-lapse derivation provides a corrected injection parameter, based on a combined test quantity and test moment performance diagram.

Description

The present invention relates to a method for estimating an injected fuel quantity, in particular an isolated injection, in an internal combustion engine having a plurality of cylinders.

The estimation of injected fuel quantities is necessary in order to be able to correctly identify the injection parameters of an injection system of an internal combustion engine and to be able to draw conclusions about the correct functioning of the injection system. The consistent and reliable injection of a requested amount of fuel is crucial to meet the new European emission regulations for motor vehicles. The undesirable emissions of internal combustion engines are due in particular to the inaccurate calibration of injection parameters in the range of small fuel masses.

Most motor vehicles have a crankshaft sensor that detects the angular velocity of the crankshaft. This variable provides an excellent source for deriving dynamic quantities derivable from individual cylinder burns. Previous technical arrangements use a high-resolution noise measurement in the engine with the aid of one or more microphones or knock sensors. These are attached to the engine block near the cylinder. According to another alternative cylinder pressure measurements are carried out with the aid of a cylinder pressure sensor. Cylinder pressure sensors may be located at various positions within the cylinder. However, both approaches have the disadvantage that they are not installed as standard in motor vehicles and therefore substantially increase the manufacturing cost of the motor vehicle.

DE 199 45 618 A1 discloses, for example, the use of a crankshaft sensor to derive the injected injection from the injection system from the rotational nonuniformity caused by combustion nonuniformity. DE 198 09 173 A1 discloses a timed fuel metering system with which small amounts of fuel are metered before the actual injection. With these small amounts of fuel, tolerances and errors noticeably affect, so that they can be taken into account in later injection operations.

Other approaches describe the adaptation of the energy supplied to the piezoelectric injectors instead of the injection timing of the injector to identify and correct the injection parameters.

The above methods have the disadvantage that they allow the verification of injection parameters only with limited accuracy and high equipment cost. It is therefore the object of the present invention to provide a more reliable compared to the prior art method, which ensures the use of the normal equipment of a motor vehicle inspection and adjustment of injection parameters.

The above object is achieved by a method according to independent claim 1. Further developments and embodiments of the present method will become apparent from the following description, the drawings and the appended claims.

The present invention discloses a method for estimating an amount of fuel injected into an engine having a plurality of cylinders, comprising the steps of injecting and combusting a test amount of fuel in a cylinder of the engine during a phase of cut off fuel supply, determining a segment time T (k) of the internal combustion engine from signals from a crankshaft sensor, calculating a generated by burning the test quantity test torque C (k) from a numerically determined second time derivative of the segment time T (k) and determining a magnitude of Test set from the calculated test torque C (k) based on a test amount test torque map.

The present method utilizes the signals provided by a crankshaft sensor, which is now standard on today's motor vehicles. With the help of known crankshaft sensors, the combustion cycles taking place in the individual cylinders of the internal combustion engine can be evaluated. According to the number of cylinders of the internal combustion engine, the total cycle of the internal combustion engine comprising 720 ° is divided into individual segments which can be used to describe the combustion in the individual cylinders. If the internal combustion engine is in a phase of disconnected fuel supply, ie the driver does not request torque via the accelerator pedal, individual test quantities are injected and ignited into individual cylinders of the internal combustion engine during coasting of the motor vehicle. Since these test quantities are small in comparison to injected fuel quantities of normal coasting phases of the motor vehicle, the burning of these test quantities does not adversely affect the driving behavior of the motor vehicle, such as, for example, jerking. Nevertheless, the injected test quantities produce by their combustion a thrust moment or test torque, which can be recognized and evaluated with the aid of the crankshaft sensor. The injection and burning of a test quantity has an immediate effect on the segment time T (k) determined by the crankshaft sensor. This change in the segment time T (k), which varies to varying degrees depending on the size of the injected test quantity, is in the present method for checking the injection parameters and thus the Functionality of the injection system used. In order to be able to use the phases of deactivated fuel supply independently of the state of motion of the motor vehicle to check the injected fuel quantity, the second time derivative of the segment time of the respectively considered cylinder of the internal combustion engine is numerically formed from the measured values of the ball-shaft sensor. In this way, for example, the influence of different engine speed ranges of the internal combustion engine is excluded, so that any phases of disconnected fuel supply can be used together to estimate the injected fuel quantity. The numerical second time derivative of the segment time T (k) represents the test torque C (k) generated by the combustion of the test amount and the amount of the torque contribution by the combustion of the test amount, respectively. Once the test torque C (k) has been determined, it is determined by means of a test quantity test torque map, which actual size of the test quantity corresponds to the determined test torque C (k). The determined on the basis of the map actual size of the test amount provides information about the extent to which the injection system of the internal combustion engine actually injects the requested amount of fuel or errors in the injection parameters are present. With this knowledge, it is possible to constantly calibrate an apparatus change of the injection system, in order to ensure in this way an optimal emission behavior of the internal combustion engine.

According to a preferred embodiment of the present invention, injecting and combusting a series of test amounts in one or a plurality of interrupted fuel supply phases of the internal combustion engine and calculating an average of the test torque C (k) from the test torques determined per injected test amount.

According to a further embodiment, the test torque is calculated as a difference between the test torque after the injection of the test quantity and the test torque before the injection of the test quantity, ie a phase of disconnected fuel supply without isolated injection.

It is further preferred to perform a recursive update of the mean value of the test torque with each further series of injected test quantities. Updating the average with each series is characterized by the fact that the measurements of several series based on few test injections together form a fuel estimate, without having to wait for a series with a much larger number of test injections or on a sufficiently long series. This increases the effectiveness of the present method compared to the prior art.

• Figur 1A enthält ein Beispiel für den Abfall der Drehzahl der Brennkraftmaschine in einer Phase mit abgeschalteter Kraftstoffzufuhr.
• Figur 1B zeigt exemplarisch die Zunahme der Segmentzeit T(k) in Abhängigkeit von der Zeit in einer Phase abgeschalteter Kraftstoffzufuhr.
• Figur 2 zeigt die aus den Messpunkten des Kurbelwellensensors numerisch berechnete zweite Ableitung der Segmentzeit T(k).
• Figur 3 zeigt ein beispielhaftes Flussdiagramm des vorliegenden Verfahrens.

The present invention and its preferred embodiments will be explained in more detail below with reference to the accompanying drawings. Show it:

• FIG. 1A shows an example of the decrease in the speed of the internal combustion engine in a phase with the fuel supply switched off.
• FIG. 1B shows by way of example the increase of the segment time T (k) as a function of time in a phase of switched-off fuel supply.
• FIG. 2 shows the second derivative of the segment time T (k) calculated numerically from the measuring points of the crankshaft sensor.
• FIG. 3 shows an exemplary flow chart of the present method.

The disclosed method is for estimating and verifying the magnitude of injected amounts of fuel injected into one or more cylinders of an internal combustion engine, respectively. In this way, it is determined whether an injection system still fulfills the assumed injection parameters, so that optimum emission values of the internal combustion engine are achieved.

As part of the process fuel test quantities or isolated injections are injected into the individual cylinders and burned in a phase off fuel supply. A phase of cut-off fuel supply of the internal combustion engine designates a period of time in which neither fuel injection is requested from the driver nor from other units of the internal combustion engine. In these phases, ideally, there is a linear decrease in rotational speed over time, as exemplified in FIG. 1A.

The linear decrease of the engine speed according to FIG. 1A corresponds to the unchanged moment of inertia of the crankshaft within this phase. If, for example, the transmission ratio G is changed via a transmission of the motor vehicle or if the crankshaft experiences disturbing forces due to poor road conditions, there is a sudden change in the moment of inertia on the crankshaft, so that the linear drop of FIG. 1A changes abruptly. Such events would normally adversely affect an evaluation algorithm based on the engine speed. However, a significant advantage of the present invention is that it is independent of the linear drop in engine speed and also less susceptible to isolated changes in speed drop.

The above identification of a phase of cut off fuel supply corresponds to the first step of the flowchart in Figure 2, which schematically shows a Embodiment of the method represents. The segment time T (k) for a multi-cylinder engine refers to the duration of one cylinder rotation when the total time for one complete cycle of the engine is divided by the number of cylinders of the engine. The segment T (k) can be determined with the aid of the signals of the crankshaft sensor of the internal combustion engine. For example, if the crankshaft sensor includes 60 teeth and the engine has four cylinders, one complete cycle of the engine is divided into four segments of 30 teeth of the crankshaft sensor. Since these teeth are detected individually by the crankshaft sensor, can be determined in this way the time for each segment in dependence on the rotational speed N of the internal combustion engine. In addition, since the sampling rate of the crankshaft sensor varies with the speed of the internal combustion engine, it is in each case adapted to a sufficient detection of the segment time T (k). If the segment time T (k) is described in seconds and NR_CYL is the number of cylinders of the internal combustion engine and N is the engine speed in Umin ^-1 , the segment time T (k) is calculated in the segment with the number k according to: $$T\left(\mathrm{Kuk}\right)=\frac{120}{N\left(k\right)\cdot \mathit{NR\_CYL}\mathrm{,}}$$

A typical course of the segment time T (k) as a function of time during a phase of cut-off fuel supply is shown in FIG. 1B. Again, the speed of the internal combustion engine during the phase shut off fuel supply decreases linearly with time, as shown by way of example in Figure 1A.

The method for estimating the quantity of injected fuel quantities or injected test quantities, which is also referred to as a characteristic value determination method (cf., FIG. 3), is based on the second numerical time derivative the segment time T (k). By directly forming the numerical derivative of the segment time T (k), one uses the computational advantage of less multiplications and divisions to arrive at a magnitude proportional to the injected test amount of fuel. In this way, rounding errors are reduced and the numerical range significantly increased when performing the calculations using fixed-point arithmetic. The final magnitude represents an average of the applied torque during the injections of the test amount produced by the force produced by the combustion of the isolated injected fuel test amount during the segment time T (k). This final quantity is referred to below as combustion statistics or test moment C (k).

Applying the numerical time derivative D [x (k)] to the function x (k) yields $$D\left[x\left(k\right)\right]=\frac{x\left(k\right)-x\left(k-1\right)}{T\left(k\right)}$$

Equation (2) contains the assumption that the actually required time term (t ₂ -t ₁ ) in the denominator according to a known time derivative experimentally corresponds approximately to the segment time T (k). This assumption considerably simplifies the further calculations. It is also conceivable to approximate the time term (t ₂ -t ₁ ) by the mean value 1/2 * (T (k) -T (k-1)).

If the second time derivative D [D [T (k)]] of the segment time T (k) is formed on the basis of equation (2), one obtains $$A\left(k\right)=D\left[D\left[T\left(k\right)\right]\right]=\frac{T\left(k\right)\cdot T\left(k-2\right)-{\mathrm{<figure-callout\; id="2"\; label="T"\; filenames="imgf0001.png"\; state="\{\{state\}\}">T</figure-callout>}}^2\left(k-1\right)}{T^2\left(k\right)\cdot T\left(k-1\right)}$$

The second derivative formation removes the local quadratic shape of the segment time data as shown in FIG. 1B. Therefore, the result of this operation is located approximately around the zero point. By doing that quadratic increase in the segment time T (k) "lost" as a function of the decrease in the speed of the internal combustion engine with the second time derivative, the dependence of the segment time T (k) is essentially removed from the speed of the internal combustion engine.

The motion of the crankshaft system is modeled here by a differential equation of second order T̈ = f (T, Ṫ). Here, the goal of estimating the amount of fuel of isolated injection is changed to estimate the resultant force experienced by the crankshaft system due to isolated injection. After completion of the evaluation of the experimental data, this force is transmitted to the size of the injected fuel quantity with the aid of a test quantity test torque map. This map was previously determined experimentally specifically for the internal combustion engine. The magnitude of this force is calculated as the norm of the differential equation f (T, T) over a short period of time after isolated injection. The corresponding formula for this calculation is $$C=\frac{1}{H}{\displaystyle \underset{t}{\overset{t+H}{\int }}}\left|f\left(T\dot{T}\right)\right|dt\mathrm{,}$$

The test torque C (k) already introduced above is discretely approximated in the context of the present method by means of a weighted linear combination of A (k). In the context of this discrete approximation, A (k) is scaled over a time interval by means of a function of the gear ratio G and the engine speed N. For the discrete approximation of the test torque C (k) is therefore obtained $$C\left(k\right)=\frac{1}{b\left(G\left(k\right).N\left(k\right)\right)}{\displaystyle \underset{i=k}{\overset{\begin{array}{cc}k-\mathit{NO}& \mathit{CYL}-1\end{array}}{\Sigma }}}a\left(j\right)A\left(j\right).a\left(j\right)\in \Re \mathrm{,}$$

FIG. 2 shows by way of example the second numerical time derivative of the segment time, which is influenced by the event of a representative isolated injection of a test amount of fuel. The calculation of the test torque T (k) is represented by the hatched area below the curve. The curve itself is formed by the function A (k) (see above). The points also represented represent sampled events during the engine cycle.

The combustion statistics or the test torque T (k) has approximately the mean value zero, if no injection and ignition of a test amount takes place in the context of the fuel cut-off phase. The variance of the test torque C (k) is estimated in phases where no injection takes place. In this way, the expected variability of the test torque C (k) is determined. In this context, essential key variables of the system are taken into account, such as the engine speed and different moments of inertia on the crankshaft. The estimated data dispersion is for detecting a system whose hardware is in a state outside an acceptable range. Furthermore, unacceptable operating conditions for the above evaluation, such as bad road conditions, can be recognized via the data scattering.

The characteristic value determination or the steps for estimating the size of the fuel quantity of an isolated injection are shown in FIG. At the same time, steps for the robustness of the estimation procedure or for the acceptance of the measurements are presented. First, in the third step of the flowchart of FIG. 3, equation (5) is applied over several segments of the engine cycle in a phase of cutoff fueling. Since in this phase initially no test quantities are injected or no isolated injection is carried out, the variance of the characteristic values can be determined as a function of the determined C (k) Determine the operating state of the internal combustion engine and / or the motor vehicle. If the variance is below a predetermined threshold, the process continues. Otherwise, the measurement is repeated or another phase of shut off fuel supply is awaited under other operating conditions of the motor vehicle and then the measurement is repeated.

Continuing the measurement, a test amount is injected into a selected cylinder as part of an injection cycle of the internal combustion engine. The injection cycle is arranged between a given number of reference cycles in which no test quantities are injected. By means of the measurement of injection cycles and reference cycles, comparison possibilities are provided in the further process. The isolated injection of a test set or a series of test set injections is made with identical control parameters for the injector to achieve comparability over a plurality of isolated injections. The corresponding test torques C (k) are determined and collected or stored according to equation (5). In the sense of the comparability already mentioned above, injection or engine cycles are alternated with injection of a test quantity with injection or engine cycles without injection of a test quantity.

With the aid of the already determined variance of C (k) of the injection cycles without test quantity injection, an expectation interval is defined. If the measurement of C (k) of the reference cycle, ie without test quantity injection, is outside the expected interval, the measured values of the following test quantity injection are not evaluated. Otherwise, a corresponding evaluation of the test quantity injection takes place. The application of the expectation interval ensures that the data from the reference cycles can actually be used to evaluate the test moments C (k). For example, drives the motor vehicle during a reference measurement by a pothole, a bad road or otherwise unpredictably changes the moment of inertia at the crankshaft, unexceptable fluctuations in C (k) of the reference cycle are generated. This prevents a reliable later evaluation.

If the C (k) of the reference measurement is in the expectation interval, the magnitude of the test torque C (k) becomes a difference of the test torque C _{after_inj} (k) after the combustion of an isolated injection and the test torque C _{before_inj} (k-NR_CYL) before combustion calculated isolated injection. This is also shown in the following equation 6. $$C\left(k\right)=C_{\mathit{after\_ini}}\left(k\right)-C_{\mathit{before\_ini}}\left(k-\mathit{NR\_CYL}\right)$$

In this way, it is not necessary to determine the force generated from the isolated injection alone. Furthermore, errors in the determination of the segment time with the help of the crankshaft sensor no longer have an effect.

In the further process, outliers are preferably removed from the collected C (k) values and an average of the collected C (k) values is made within each series of isolated injections. These steps improve the accuracy and robustness of the final estimate of the actual amount of fuel injected. For each series of isolated injections, an average and a variance of the results are calculated. Using this computed data, the outliers are removed based on the assumption that the data dispersion obeys a Gaussian distribution. The mean C _i of each series i and the variance var (C) _i of each series I is calculated $$\widetilde{c_i}=\frac{1}{n_i}{\displaystyle \underset{k}{\Sigma }}C\left(k\right).\mathrm{var}\left(C\right)=\left({\displaystyle \underset{k}{\Sigma }}{C}^2\left(k\right)-n_i{\widetilde{C_i}}^2\right)/\left(n_i-1\right)$$

Each series of injected test quantities or isolated injections can be determined in different phases of cut off fuel supply. The averages of each series are collected until a sufficient number n _T of evaluated injection events have been collected. The number of injection events is sufficient if a reliable estimation of the test torque generated by the isolated injections is possible with respect to the injection parameters which are the same everywhere. Of course, the accuracy of this estimation also affects the later determination of the actually injected test set based on the test set test torque map.

The final mean test moment C and the variance var (C) of the test results are recursively updated with each series i of isolated injections. This is shown in the following equations (8). $$\begin{array}{c}\tilde{C}=\frac{n_T\tilde{C}+n_i\widetilde{C_i}}{n_T+n_i}.\mathrm{var}\left(C\right)=\frac{\left(n_T-1\right)\mathrm{var}\left(C\right)+\left(n_i-1\right)\mathrm{var}\left(C_i\right)}{n_T+n_i-1}\\ n_T=n_T+n_i\end{array}$$

The essential advantage of the present invention is that, despite the sporadic repeatability and duration of the periods of fuel cut-off, results are achieved with high accuracy and low susceptibility to external disturbances and changes in boundary conditions. The averaging over a plurality of series of isolated injections and the recursive updating of the specific test moments makes it possible for even a slight change in the injection conditions to be taken into account in the control of the injection parameters for a great variety of reasons. On this basis, it is ensured that strict emission requirements are met.

Based on the averaged and recursively updated test torques, the actually injected quantities of the test quantities are derived from the test quantity test torque characteristic field. The knowledge of the actual quantities of the test quantities then in turn makes it possible to calibrate the control parameters, for example an injection system, and to adapt them to the requirements of the respective internal combustion engine.

Claims (4)

A method of estimating an injected amount of fuel into a multi-cylinder internal combustion engine, comprising the steps of: a. Injecting and combusting a test amount of fuel in a cylinder of the internal combustion engine during a phase of cut off fuel supply, b. Determining a segment time T (k) of the internal combustion engine from signals of a crankshaft sensor, c. Calculating a test torque C (k) generated by the burning of the test quantity from a numerically determined second time derivative of the segment time T (k) and d. Determining a size of the test set from the calculated test moment C (k) based on a test set test moment map. Method according to claim 1, with the further step: Injecting and burning a series of test amounts in one or a plurality of phases of interrupted fuel supply of the internal combustion engine and Calculating an average of the test torque C (k) from the test torques determined per injected test amount. Method according to one of the preceding claims, with the further step: Calculating the test torque calculated as a difference from the test torque after injecting the test amount and the test torque before injecting the test amount. Method according to claim 2 or 3, with the further step: Recursively updating the average of the test moment with each additional series of injected test sets.

Images