EP0150437B1 - Measuring system for the fuel-air mixture in a combustion engine - Google Patents

Description

State of the art

The invention relates to a device for fuel-air mixture metering for an internal combustion engine according to the preamble of the main claim. Methods are already known which regulate the mixture composition to the minimum consumption in the part-load range of an internal combustion engine. In the case of extreme value regulations, it has already been proposed for this purpose to sweep the air quantity supplied to the internal combustion engine by means of a test signal. Due to the relatively large distance between the bypass and the throttle valve and the individual cylinders, run times are thereby created which limit the wobble frequency and result in a relatively slow control behavior. Furthermore, a complex actuator, for example an air flap in an air bypass, is required.

In order to avoid these disadvantages, for example in DE-A 2 941 977 for an extreme value control on minimum specific fuel consumption, it was proposed to vary the metered fuel quantity as a test signal and to determine the operating point of the minimum fuel consumption of the internal combustion engine via the maximum efficiency. The maximum efficiency is determined on the basis of the quantities torque of the internal combustion engine and fuel metering signal by division. However, this method also requires an intrinsically complex torque sensor and at least one computing unit to carry out the division.

DE-A 2 507 055 discloses a device for fuel / air mixture metering for an internal combustion engine with an extreme value control which intervenes in the fuel metering by means of a test signal generator for regulating a minimum specific fuel consumption, in which the engine speed and a fuel metering signal are as actual information can be used to regulate the minimum specific consumption value. The aim is to optimize consumption by wobbling the amount of air or to optimize performance by wobbling the amount of fuel.

Advantages of the invention

With the device according to the invention for fuel-air mixture metering for an internal combustion engine with the characterizing features of the main claim, on the other hand, extreme value control for minimal specific fuel consumption can be carried out without additional sensors.

Only the variables speed n and injection time t are used as input information.

In particular, it proves to be advantageous to use information about the gearbox of the gearbox connected to the internal combustion engine for control purposes.

Further advantages of the invention and expedient refinements of the invention result in connection with the subclaims from the following description of the exemplary embodiment.

drawing

• Figur 1 mittlerer Nutzdruck p, einer Brennkraftmaschine aufgetragen über der Luftzahl mit dem Parameter Kraftstoff- bzw. Luftmenge,
• Figur 2 Zusammenhang zwischen Luft- bzw. Kraftstoffmenge und der Luftzahl \ für einen vorgegebenen konstanten Mitteldruck p_"
• Figur 3 Blockschaltbild einer ersten Ausführungsform mit Extremwertregelung,
• Figur 4 Prinzipdarstellung einer Extremwertregelung,
• Figur 5 Amplitude und Phasenlage eines Bandpasses in einer Extremwertregelung,
• Figur 6a der Vorsteuerung überlagerte Regelung mit multiplikativem bzw. additivem Eingriff,
• Figur 6b der Vorsteuerung überlagerte Regelung zur individuellen Kennfeldanpassung,
• Figur 7a Anpassung von Kennfeldeinzelwerten,
• Figur 7b Anpassung von Kennfeldbereichen,
• Figur 7c multiplikative Anpassung des gesamten Kennfeldes,
• Figur 8 Kennfeldlernverfahren,
• Figur 9 Kennfeldausschnitte mit Stützstellen,
• Figur 10 Kennfeldlernverfahren mit Mittelwertbildung,
• Figur 11 Blockschaltbild einer dritten Ausführungsform,
• Figur 12 α-n-Kennfeld für die Einspritzzeit t_;,
• Figur 13 Schaltungsaufbau für eine a-n-Gemischvorsteuerung mit einer additiven Regelung,
• Figur 14a Dremoment einer Brennkraftmaschine aufgetragen über der Einspritzzeit für konstante Drehzahl n und Luftmenge Q_L,
• Figur 14b Wirkungsgrad bzw. spezifischer Kraftstoffverbrauch aufgetragen über der Einspritzzeit für konstante Drehzahl n und Luftmenge Q_L,
• Figur 15 Blockschaltbild einer vierten Ausführungsform,
• Figur 16 a-n-Kennfeldausschnitt für die Einspritzzeit,
• Figur 17 Blockschaltbild einer fünften Ausführungsform.

Embodiments of the invention are shown in the drawing and explained in more detail in the following description. Show it :

• FIG. 1 average useful pressure p of an internal combustion engine plotted against the air ratio with the parameter fuel or air quantity,
• FIG. 2 Relationship between air or fuel quantity and the air ratio \ for a predetermined constant mean pressure p _"
• FIG. 3 block diagram of a first embodiment with extreme value control,
• FIG. 4 basic illustration of an extreme value control,
• FIG. 5 amplitude and phase position of a bandpass in an extreme value control,
• FIG. 6a the control superimposed on the pilot control with multiplicative or additive intervention,
• FIG. 6b the control superimposed on the pilot control for individual map adaptation,
• FIG. 7a adaptation of individual map values,
• FIG. 7b adaptation of map areas,
• FIG. 7c multiplicative adaptation of the entire map,
• FIG. 8 map learning method,
• FIG. 9 map sections with support points,
• FIG. 10 map learning method with averaging,
• FIG. 11 block diagram of a third embodiment,
• Figure 12 α-n map for the injection time t _; ,
• FIG. 13 circuit structure for a mixture pre-control with an additive control,
• FIG. 14 a torque of an internal combustion engine plotted against the injection time for constant speed n and air quantity Q _L ,
• FIG. 14 b efficiency or specific fuel consumption plotted against the injection time for constant speed n and air quantity Q _L ,
• FIG. 15 block diagram of a fourth embodiment,
• FIG. 16 an map section for the injection time,
• FIG. 17 block diagram of a fifth embodiment.

Description of the embodiments

The design of devices for air / fuel mixture metering for internal combustion engines is generally based on the following requirements: achieving a minimum specific fuel consumption, compliance with low exhaust gas emissions and ensuring satisfactory driving behavior.

Various control methods, which are partially described below, are generally used, which enables the use of simple and inexpensive sensors and actuators, guarantees freedom from maintenance and increases long-term stability. Variations of specimens can also be neglected, interchangeability of, for example, encoders is ensured and the facility can be adapted more easily to different motor types. The use of controls also results in functional improvements such as the optimization of the operating behavior of the internal combustion engine in the start, warm-up and idle phases and in the full-load range. The same applies to transient phases of the internal combustion engine, for example during acceleration or overrun.

In contrast to a regulated system, in which disturbances that may occur are recorded, but the adaptation of the internal combustion engine to the new conditions is very slow due to uneven combustion processes, gas runtimes, etc., control systems allow a very fast adaptation to changed input conditions. On the other hand, the disturbance variables can only be considered incompletely or with considerable effort. By using a self-adapting map, the map providing pilot values that are influenced by a superimposed control, the respective advantages of controlled and regulated systems are to be used.

For a brief explanation of the control methods, characteristic curve fields of an Otto engine are shown in FIG. 1 and FIG. In Figure 1, the average useful pressure proportional to the power is plotted against the air ratio with the fuel quantity (dashed lines) or the air quantity (solid lines) as parameters. From these characteristic curves it follows that a predetermined average useful pressure or a predetermined output (here an average useful pressure p _e = 5 bar) can be realized within the predetermined limits with any desired λ. The smallest amount of fuel is required with an air ratio of slightly less than λ = 1.1. This results from the fact that the curves for constant fuel quantity have a maximum in the range A = 1.1. In contrast, there is the maximum power for the curves of constant air volume with lambda values in the region of X = 0.9. In the first case, namely for a predetermined constant fuel quantity, a maximum output of the internal combustion engine is achieved if the air quantity is metered in such a way that the air-fuel mixture assumes a lambda value of λ = 1.1. If the air is tracked in a fuel-guided injection system in such a way that a maximum output results, the internal combustion engine is automatically operated in the range of a minimum specific fuel consumption.

In the second case, in which the internal combustion engine is operated with a constant air quantity for a maximum power output at X = 0.9, there is operation at maximum power. This relationship is clear from FIG. 2, in which the air or fuel quantity to be metered as a function of the lambda value is plotted for a predetermined, constant average useful pressure. This average useful pressure is achieved with a minimum of fuel if the lambda value of the air-fuel mixture is λ = 1.1. This point is therefore identical to the minimum specific fuel consumption be _min - On the other hand, the minimum amount of air with which this average useful pressure can be achieved includes an air-fuel mixture with values of X = 0.9. Here there is therefore a maximum power output P _{max of} the internal combustion engine for a given air quantity.

Based on these relationships, the following control methods are available for metering the mixture in an internal combustion engine: In the entire part-load range, control is based on a minimum specific fuel consumption of the internal combustion engine, that is, on a maximum of the curves shown in dashed lines in FIG. 1 (be _{min control} ). In the case of full load, on the other hand, the system regulates to a maximum power, ie to a maximum of the curves shown in solid lines in FIG. 1 (P _{max control} ). Since in both cases the setpoint is given by a maximum of the power output of the internal combustion engine for a given amount of fuel or air, an extreme value control is appropriate. However, a lambda map control is also conceivable, with which the corresponding lambda values of the air-fuel mixture are specified as a function of the power output of the internal combustion engine.

Control systems for an internal combustion engine, such as a lambda control, a knock control or ignition timing control, can react relatively slowly to disturbance variables due to the existing dead times or running times. It has therefore proven to be extremely advantageous to use a pilot control for the fast and dynamic processes within an internal combustion engine. The superimposed control can intervene, for example, multiplicatively or additively on these pilot control values. Through the use of modern electronic means, such as memory and microcompu It is also possible to implement the feedforward control by means of a map, the map values of which can be addressed, for example, as a function of the speed and the load of the internal combustion engine. The superimposed control can then, on the one hand, influence the read map values multiplicatively or additively without changing the map values stored in the memory. On the other hand, however, it is also possible to change the map values per se by means of the superimposed control. If the influence of disturbance variables is constantly taken into account by changing map values, one speaks of a self-adapting or also of a learning map. As will be shown below, a combination of the two methods described last can also be very advantageous.

The basic structure of the system uses a map, the input variables of which in the simplest case are the speed n and the throttle valve position a. When initializing for the first time, relatively rough initial values are loaded into this map. An ongoing adjustment then takes place during operation. An essential concept is to subdivide the map into different areas, for example idling, part-load, full-load and thrust areas. With the exception of push operation, a specific control concept is provided in each area, which adapts this area to the respective requirements, so that a "learning" map is available. If the motor vehicle is taken out of operation, it is possible to save the map learned last and to use it as the initial map again when starting again.

FIG. 3 shows the block diagram of a first embodiment of the overall system. The fuel quantity to be metered to the internal combustion engine is controlled via a characteristic diagram 20, to which the speed n and the throttle valve position a of a throttle valve 21 are fed as input variables. The throttle valve is actuated by an accelerator pedal 22. The injection time t stored in the map 20 _; is converted into a corresponding fuel quantity Q _K via an injection valve 23. This amount of fuel Q _K , as well as the amount of air Q _L determined by the throttle valve position, are fed to a symbolically illustrated internal combustion engine 24, a certain torque M being effected as a function of the lambda value of the air / fuel mixture. The controlled system “internal combustion engine” can be roughly approximated schematically by an integrator 25. The output variable n of the internal combustion engine is used to control the characteristic diagram 20. This part of the overall system described so far is a pure control of the mixture composition.

In this exemplary embodiment, the superimposed control is based on extreme value control. For this purpose, depending on the control method (see the explanations below), either the air quantity Q _L, for example via a bypass with a stroke ΔQ _L, or the injection time t _; wobbled with a stroke Δt. The test signals required for this purpose are generated by a test signal generator 26. This test signal generator acts on the amount of fuel or air, depending on the control method, the wobble frequency being able to be chosen to be constant or depending on the speed. The torque changes of the internal combustion engine generated by the test signal have an effect on changes in speed, so that these changes in torque can be analyzed by a measuring device 27 which receives signals proportional to the speed. The measuring device 27 consists of a preferably digital filter 28 and a subsequent evaluation unit 29 which examines the filtered signal in terms of magnitude and / or phase and compares it with the output signals of the test signal generator 26. It has proven to be advantageous to construct the filter 28 using digital technology. It works discretely, whereby the sampling rate can be either in a fixed time grid or proportional to the speed. Since the filter is precisely matched to the sweep frequency, interference signals can be largely suppressed. A controller 30 compares preferably the phase position of the filter output signal and a phase setpoint, the difference between these two signals being fed to an integrator 31, which in the simplest case can be designed as an up-down counter. On the one hand, the output signal of this integrator 31 can be used for a multiplicative influencing of the characteristic diagram. As will be shown further below, characteristic mapping processes can also be useful, in which individual areas of the characteristic diagram are specifically adapted. Such methods are shown schematically by block 32.

The principle representation of an extreme value control of FIG. 4 serves to explain the functioning of the system of FIG. 3.

4 shows the average useful pressure p _e as a function of the lambda value of the air-fuel mixture. A test signal is superimposed on the input variable, the air-fuel mixture with a predetermined lambda value, which either occurs sporadically and, for example, takes the form of a staircase function or occurs periodically and has a sine or rectangular shape. The reaction of the internal combustion engine to these test signals can be detected via the change in the average useful pressure p _e , but particularly advantageously via the change in torque or the associated change in speed. As can be seen from FIG. 4, either the change in amplitude of the mean useful pressure (or the torque or the speed) or the phase of this output variable in relation to the phase of the test signals is suitable as the variable to be analyzed.

In the be _{min control,} the test signal is superimposed on the input variable by air wobble, for example via a bypass. This control method is used in the exemplary embodiment in FIG. 3.

A rough precontrol of the lambda value of the air / fuel mixture is specified via the throttle valve 21 and via the on map 20 for the injection time. The superimposed control comprises a test signal generator 26, a measuring device 27 evaluating the speed changes and a controller 30 which influences the characteristic diagram 20. The amount of air to be metered wobbles around AO _L. FIG. 3 is to be understood in such a way that the signals of the test signal generator 26 influence the amount of air to be metered. The reaction of the internal combustion engine 24 to this wobbling of the air-fuel mixture supplied can be analyzed by changes in the rotational speed. For this purpose, a measuring device 27 is used which, in the present special case, consists of a digital filter 28 for suppressing interference frequencies and an evaluation unit 29 which evaluates the changes in speed with regard to the amount and the phase. The output variable of this measuring device 27 is compared as the actual value of the speed changes with the target value Δn = 0 of the speed changes typical for extreme value control, the difference between the actual and target values via blocks 31 and 32 on the characteristic diagram 20 to different, further below acts in an explanatory manner.

In FIG. 5 the output signal of the bandpass is plotted to clarify the functioning of the evaluation unit 29, namely the amplitude as a function of lambda in the upper part and the phase position for two lambda values above and below the ideal value, the be _min point, in the lower part , on which the figure 4 was specialized. The conditions would result in the same way for a regulation to power _maximum P _max , only the lambda value being in the rich range. The output amplitude of the bandpass is a measure of the size of the speed changes. According to the representations in FIG. 4, the change in the output amplitude of the bandpass assumes the value 0 at the extreme value. Deviating from the optimal value on both sides, the amplitude increases steadily. However, the value of the amplitude alone does not say on which side of the extreme value one lies. The extreme value is therefore determined by evaluating the phase of the output signal of the filter 28. It would also be possible to use the change in amplitude as a measured variable.

In the lower part of FIG. 5, on the one hand, a test signal of an arbitrary, here rectangular shape, and in comparison the output variable of the filter is plotted. Depending on whether the lambda value of the fuel mixture is above or below the bemm point, the output signal of the filter has a different phase shift with respect to the test signal. Depending on the phase position, this can be used to clearly determine whether the mixture is too rich or too lean in relation to the be _min point.

In the controller 30 of FIG. 3, a comparison now takes place between the phase position of the output signal of the filter 28 and a phase setpoint for the be _min point. The difference between these two signals is integrated in the simplest case, for which purpose, for example, an up-down counter can be used in a digital embodiment. The counter reading corresponds to a factor by which the injection map is multiplied or a certain map area is modified. With the be _{min control} , the air has to be wobbled, so that due to the large distance between the bypass at the throttle valve, with which the amount of air is wobbled, and the cylinders, run times are created by which the wobble frequency is limited. Due to the presence of vehicle-specific resonance frequencies, the phase setpoint for the be _min point can be shifted depending on the speed and possibly also on the load.

In particular for the exemplary embodiment in FIG. 3, in which an air bypass is provided for wobbling the air quantity supplied, an idle charge control can advantageously be used, with which the LL speed of the internal combustion engine is independent of load changes, such as, for example, by switching on the air conditioning system or similar are caused, is kept constant. Such an idle charge control is known for example from the applicant's German published application DE-OS 3 120 667.

In the following, the principle of adapting characteristic maps, as is already known for injection systems, carburetor systems and also ignition systems, will now be discussed in more detail. A rough classification of the map adaptation method can be specified as follows: FIG. 6 shows a possibility in which the map values for a pre-control of the injection time remain unchanged, but multiplicative or additive corrections can be applied to the output values of the map via the superimposed control. However, the map values themselves cannot be changed by the overlaid control. The advantage of this method is that the method is very simple and inexpensive to carry out. The disadvantage, however, is that the structure of a map that has been predetermined once is no longer modified.

In contrast, FIG. 6 b shows a map learning method in which the individual map values of the map are continuously adapted by the superimposed control. More precisely, this means that at each operating point specified by the input variables, the associated map output variable is adapted to the respective optimum value by a control method. When leaving the respective operating point, the last determined output variable is saved and remains unchanged until this operating point is selected again. An advantage of this method is that the map can be adapted to any structure. Not so advantageous is the fact that to change the entire map, all map output sizes have to be controlled individually. However, this is not always guaranteed, because on the one hand different operating points, very much are controlled rarely or never and, on the other hand, since the length of stay in the individual operating points is often so short that no adjustment can be made.

The disadvantages of both methods can be solved in an advantageous manner by a compromise that lies between these two extreme possibilities. In addition to the directly controlled output variable, an area around this variable is influenced, this influencing of adjacent characteristic map values decreasing with increasing distance from the respective output variable. It has been found to be particularly advantageous in this compromise that an almost arbitrary adaptation of the characteristic diagram is possible and that areas that are never or only rarely controlled are also influenced.

The adaptation methods explained above are to be illustrated on the basis of FIG. 7, which shows a section through a histogram-like actual value map with the corresponding target values identified by a solid line. The adaptation of individual values is shown in FIG. 7a, the controlled output variable being identified by an arrow. Although this individual value is correctly adjusted by the control system according to the setpoint curve, the structure of the actual value map curve can only track the setpoint after all map values have been approached. When leaving the controlled output variable and transitioning to a closely adjacent map size, this must be adjusted in a similar direction to the previous values. The other extreme case, namely that of a multiplicative adaptation of the overall map, is shown in FIG. 7c. A factor is obtained from the deviation of the map size indicated by an arrow from the target value, which adapts the corresponding map value correctly, but changes all other map values in the same sense. As can be seen from the selected setpoint curve, the desired setpoint curve of the map cannot be exactly achieved with such a multiplicative adjustment. There are various options for adapting a mixed form of the two methods, as is shown schematically with reference to FIG. 7b. One possibility is to subdivide the map into support points. In the simplest case, intermediate values are calculated, for example, by linear interpolation. When the map is adapted to the corresponding target value, only the support points are changed, so that the surrounding areas are adjusted by the interpolation. In this case, the environment of the changed support point value is automatically changed in the same sense as the support point itself, but weighted less with increasing distance from the support point. With this map learning method, it is not necessary to approach each individual map size in order to change it. This means that, on the one hand, the characteristic map can be adapted quickly and, on the other hand, any given structure can be adapted at least approximately.

Due to the system-related runtimes, the control factor is not always constant, even in stationary operation, but shows time fluctuations. For this reason, an averaging of the control factor is appropriate, in which case these averaged control factors are only incorporated into the characteristic diagram at predetermined times. After the training, the control factor is reset to one. This measure has the advantage that the map can be adapted safely, although the time period of the adaptation can be extended.

FIG. 8 should be used to explain the advantages of such averaging. For reasons of simplicity, only three interpolation point values S1, S2, S3 have been plotted here, which also assume the same values. The bold actual value characteristic curve therefore assumes the shape of a straight line. In the present example, the setpoint characteristic curve shown in broken lines deviates considerably from the course of the actual value characteristic. A support point intake area is defined around each of the support points, which in the present special case corresponds to half the distance between two support points, as indicated for support point S2 in the drawing. Each support point can only be changed if one or more working points in the catchment area of this support point are activated. If, for example, the operating point I is activated for a longer period of time, then the desired and actual value (assuming a linear interpolation) can only be reached in this operating point if the interpolation point value S2 is raised from its initial value E to the new value A. If, on the other hand, one starts from the working point II, the interpolation point value S2 must be raised to the value D, so that the desired and actual values match for the working point II. In both cases the base did not assume its correct value, which should be B. On the one hand, this clear illustration shows that the adaptation is better the closer the working point is to the support point, and on the other hand that with a single working point in the catchment area of the respective support point, it cannot always be adapted exactly.

However, it is possible not to influence the support point immediately, but to average the correction values as long as the working point is in the catchment area. When leaving the area, the support point is corrected with this mean value. In the present example, point C would result for support point S2 according to this procedure. Although this value does not exactly correspond to the target value B, it is already quite close to the target value point. If further operating points are approached within the catchment area of the respective support point, the continued averaging of the calculated values results in a continuous approximation of the actual support point value to the reference point setpoint.

FIG. 9 shows a section from any map. The input variables, in In the present case, the speed n and the throttle valve position a are quantized and each combination of these input variables is an output variable, here the injection time t _; assigned. In the hardware version, the output variables are stored in a read-write memory, the input variables each determining the address within the memory. In the present case, a map with 3 x 3 support points was chosen as a simple example, which are marked with dots in the figure. A linear interpolation can be used to calculate three intermediate values between two interpolation points, so that a total of 81 map values result for the special case selected here.

The previously described averaging of the correction values within a region of the reference point is to be illustrated on the basis of FIG. In the upper part of this figure, nine support points (3 x 3) are shown in sections, the catchment area of a support point being shown hatched. The driving curve, which is given by the temporal change in the input variables of the map, here throttle valve position a and speed n, is shown as a solid line. This traveling curve in the entry region of the selected reference point occurs at the point A at the time t _a and exits after a certain time this supporting point portion at point B at time t _b.

The lower part of FIG. 10 shows the clear course of the control factor (solid line) in the period between t _a and t _b and the time-averaged control factor (dashed line). The averaging process now proceeds as follows. If the driving curve changes from the catchment area of a support point to another catchment area (at time t _a , t _b ), then the support point of the catchment area just left is adjusted and the control factor is reset to the neutral value one. The control factor is averaged at the point in time at which the driving curve lies in the catchment area of a support point. It can prove to be advantageous that the averaging only starts after a certain number of revolutions of the internal combustion engine (for example 16 revolutions). As a result, overshoots can be excluded and, on the other hand, a distinction can be made between dynamic and stationary driving of the internal combustion engine. A first-order digital low-pass filter, in particular, is used for the averaging. If the driving curve leaves the relevant catchment area, this averaged value is wholly or possibly only partially incorporated into the support point and then the control factor is set to the neutral value one.

A characteristic of this learning process is the fact that the properties of the existing control loop remain unchanged. The control factor continues to directly influence the manipulated variable within a reference point environment. Only after a clear change tendency has been detected by averaging several correction values within a base area, is the change incorporated into the associated base when leaving this base area. The interpolation process results in a jump in the manipulated variable, which, however, does not have a disruptive effect. It can prove useful to use a calculation method to reset the control factor in such a way that a jump is avoided.

A change limit, which uses the original state of the map as a reference, ensures that even in the event of a malfunction, an "operational" map is always retained. At the same time, a warning can be issued when the limitation is activated, since there is a high probability that there is a serious defect within the control loop or on the motor. The map in the original state also enables a comfortable emergency operation function.

In FIG. 11, which is identical to the systems of FIGS. 2 and 5 with respect to the pilot control of the mixture composition, an embodiment of the Kennfeidier method with averaging was shown in block diagram form. Although in this case the control superimposed on the precontrol is designed as an extreme value control, the principle of the mapping method presented here remains unaffected. It would also be possible to use the lambda control ((λ = 1) control, lean control or the like) presented in FIG. 5 instead of extreme value control. In any case, the output signals of the measuring device 27 of whatever type are fed to the controller 30. A comparator 40, in which the actual setpoint value comparison is carried out, is used to control a component 41, which is preferably designed as an integrator in the present exemplary embodiment. The output signals of this component 41 have a multiplicative influence on the output variable t _; of the map 20. On the other hand, they serve to control an average value generator 42, which in turn influences the individual map or reference point values of the map 20 on the output side. The connection between the mean value image 42 and the characteristic diagram 20 can be interrupted via a switch S1. Furthermore, the two blocks mean value images 42 and the component 41, which is preferably designed as an integrator, can be set to predetermined initial values A _o and B _o via further switches S2 and S3. The switches S1, S2 and S3 are controlled by a range detection 43, to which the throttle valve position a and the rotational speed n of the internal combustion engine are supplied as input variables.

At this point it should be emphasized once again that the parameters throttle valve position a and speed n for characterizing the operating state of the internal combustion engine have an exemplary character. Other parameters, such as B. the intake manifold pressure, the amount of air, the air mass or the exhaust gas temperature are conceivable as input variables.

As already described in the text for FIG. 10, a catchment area is defined for each support point. As long as the driving curve of the internal combustion engine lies within a catchment area, the correction factor is averaged in the mean value diagram 42, possibly after a speed-dependent delay time, for example, but the characteristic diagram is not influenced. The value output from the characteristic diagram 20 is permanently influenced by the output signal of the controller 30.

If the driving curve leaves the relevant catchment area of the support point, the area detection detects this and actuates the three switches S1, S2 and S3. Using switch S1, the averaged correction value can be incorporated into the last point to be controlled. In addition, the switches S2 and S3 reset the mean value images 42 and the component 41 to initial values, namely A _o and B _o . This learning process can be carried out in the same way for the next support point that has been approached.

In addition, a map for the injection times t is shown in FIG. 12 _; (in milliseconds), the throttle valve position a (in degrees) and the speed of the internal combustion engine (in revolutions per minute) serving as input variables. The map consists of 8 x 8 support points, namely eight speed and eight throttle valve positions. The 64 values for the output variable t _; are stored, for example, in a read-write memory and can be changed using the control methods already described above (be _min -, P max control) in the corresponding areas identified by different hatchings. For small throttle valve angles and speeds below approx. 1,000 revolutions per minute, the speed is regulated by means of an idle control with a higher-level bemin control. For higher engine speeds with the throttle valve almost closed, the engine is switched off from overrun. Over a wide, unshaded area, the part-load area, a _min- regulation of the mixture supply for the internal combustion engine is appropriate. In contrast, particularly when the throttle valve is fully open or the throttle valve is almost fully open and the internal combustion engine speeds are low, regulation to maximum power, P _max regulation, is advisable. These different control methods can be carried out, for example, with an arrangement as shown schematically in FIG. 3.

Furthermore, various enrichment functions, such. B. the warm-up or acceleration enrichment is provided. During the warm-up enrichment, the mixture is enriched by means of a temperature-dependent warm-up characteristic curve, whereby the characteristic diagram itself remains unaffected. In contrast, during the acceleration enrichment, a temporary change in the wetting of the intake manifold wall must be compensated. The short-term mismatch resulting from this can be corrected by increasing the fuel quantity by a factor which corresponds to the change in the position of the throttle valve over time. Because the throttle valve position is used as an input variable for the acceleration enrichment, this enrichment responds very quickly.

The hardware circuitry for carrying out an a-n mixture pre-control with a superimposed, adaptive control by means of a microcomputer (for example INTEL 8051) and the associated peripherals is shown schematically in FIG. In a microcomputer 50, the components CPU 51, ROM 52, RAM 53, timer 54, first input / output unit 55 and second input / output unit 56 are connected to one another via an address and a data bus 57. An oscillator 58 is used to time the program sequence in the microcomputer 50, which is connected on the one hand directly to the CPU 51 and on the other hand via a divider 59 to the timer 54. The signals of an exhaust gas probe 63 of a speed sensor 64 and of a reference mark sensor 65 are fed to the first input / output unit 55 via conditioning circuits 60, 61 and 62. The battery voltage 66, the throttle valve position 67, the cooling water temperature 68 and the output signal of a torque transmitter 69 serve as further input variables, which are supplied to a series circuit comprising a multiplexer 74 and an analog-digital converter 75 via associated conditioning circuits 70, 71, 72 and 73. The outputs of the analog-digital converter 75 are connected to the bus 57. The function of the multiplexer 74 and the analog-to-digital converter 75 can be implemented, for example, by the 0809 module from National Semiconducters. The multiplexer 74 is controlled via a line 76 starting from the first input / output unit 55. The second input / output unit 56 controls an air bypass 79 and injection valves 80 via power output stages 77 and 78. Further output signals of this input / output unit 56 can be used for diagnostic purposes or for ignition control.

Not all input and output variables shown here are absolutely necessary for each of the control methods described so far. The exhaust gas probe 63, the conditioning circuit 60, the torque transmitter 69 and the conditioning circuit 73 can be omitted for extreme value control to minimize fuel consumption by wobbling the air bypass 79 or the amount of fuel (injection valves 80). If the air ratio lambda is controlled instead of this extreme value control, the torque transmitter 69, the conditioning circuit 73, the output stage 77 and the air bypass 79 can be dispensed with. The torque transmitter 69 including the conditioning circuit 73 is necessary for a modified control method to be described.

The program flow for the example of an extreme value control according to FIG. 3 is explained in more detail using the following flow chart. (The other control methods already described or still to be described can be implemented in a simple manner with changed input variables and changes in the program structure, which pose no difficulties for the relevant specialist):

After this block diagram representation of the program sequence for extreme value control, some further developments or improvements and simplifications of the control methods described so far are to be discussed.

As was already shown in the description with reference to FIGS. 1, 2 and 3, the extreme value control to the _minimum consumption be _min requires wobbling of the air, for example via an air bypass, which bridges the throttle valve. The air mixture requires a certain amount of time to travel through the relatively long distance between the bypass and the individual cylinders, the frequency of the air wobble being limited on account of these transit times and the regulation thus becoming relatively slow. In contrast, the amount of fuel can be wobbled at a relatively high frequency, since the injection valves are arranged directly on the combustion chamber and runtime effects are thereby negligible. Furthermore, various methods are now to be disclosed in which a control to a consumption minimum can be implemented as a test signal using the fuel wobble. An additional advantage of this method results from the elimination of the air bypass.

To explain the basic idea, FIG. 14 shows the torque M of an internal combustion engine plotted over the uncorrected injection time t in FIG. 14a, and FIG. 14b shows the efficiency or the specific fuel consumption plotted over the uncorrected injection time t _e . The curve of the torque shown in FIG. 14a at constant air volume and speed can be derived from the solid lines in FIGS. 1, 2, the injection time serving as the abscissa instead of the lambda value of the mixture. Since the quotient of torque M and injection time t corresponds to the efficiency, the drawn tangent m indicates the maximum of the efficiency or the minimum of the specific fuel consumption. The corresponding curves for efficiency and specific fuel consumption are shown in FIG. 14b.

It is now advisable to sweep the injection time t and to determine the respective torque and the efficiency η ~ M / te of the internal combustion engine with the aid of a torque transmitter, as shown in FIG. 13 with the number 69. If this value is filtered with a digital bandpass filter, for example, and compared with the test signal, it can be determined from the phase position of the test signal and the signal at the output of the bandpass filter (see also description of FIGS. 3, 4, 5) whether the basic adjustment is on the right or to the left of the maximum. Corresponding corrective interventions are then possible via a controller. Since wobbling the injection time causes maximum changes in torque, the wobble stroke must be kept small in practical driving. It should be noted that the torque measurement is received as an absolute measurement. A shift in the zero point, for example, by means of offset voltages immediately means a shift in the calculated maximum. It proves advantageous that the air bypass for wobbling the amount of air sucked in can be omitted for this control method. Basically, this principle of wobbling the injection time can also be used for other mixture metering systems that do not necessarily derive their input variables from the speed and throttle valve position.

However, the invention takes a different approach.

• mit
• M = Drehmoment
• W = Lastmoment
• AM = Mittelwert der Momentänderung über eine Umdrehung
• 8 = Trägheitsmoment
• T = Periodendauer einer Umdrehung
• ΔT = Änderung der Periodendauer.

A method according to the invention for regulating the consumption minimum is described below, in which the fuel quantity is swept as a test signal, but no torque transmitter is required. It is possible, as the following equation shows, that the torque can also be determined from the change in speed:

• With
• M = torque
• W = load torque
• AM = mean value of the torque change over one revolution
• 8 = moment of inertia
• T = period of a revolution
• ΔT = change in period.

If the quotient ΔM / Δt _e is now formed, the slope of the torque curve can be determined according to FIG. 14a. If, on the other hand, you have measured the slope for the be _min point in the individual operating points of the internal combustion engine and stored it as a setpoint in a memory, then a control can be set up using a setpoint / actual value comparison. However, with this method it is also possible to regulate to operating points by specifying other setpoints that do not correspond to the bemin value.

As can be seen from the equation, the moment of inertia 6 is included in the calculation of the slope. However, this changes depending on the gearbox selected and the load on the internal combustion engine. The influence on the calculated slope is generally very small in vehicles with torque converters. The influence cannot always be neglected in vehicles with a manual transmission. Here, for example, it is advisable to specify the setpoints depending on the gear or charge. A simple possibility is to determine the injection map only in one gear, for example the highest gear, and as given for the other gear gears presuppose. Although the equation applies exactly only on the assumption of a constant load torque W, the error resulting from small changes in the load torque can be neglected to a good approximation under normal operating conditions of the internal combustion engine.

A simplification or improvement of the described injection method with map pre-control and a superimposed control, with different control methods depending on the operating range of the internal combustion engine, results from the arrangement described below. The map is, as already described in Figure 12, depending on the input variables such. B. the throttle valve position a and the speed n divided into different areas, namely idling, thrust, part load and full load. The design is also based on the intention of avoiding wobbling of the air volume for the _{min control} in the part-load range. For this purpose, the map values of the precontrol in the full load range are adjusted so that the engine works at maximum power. The air ratio is then, as for the idle range, at values As1. In the partial load range, the map values are adapted to the consumption minimum be _{m; n} . The air ratio A varies here between values 1.1 ≤ λ≤ 1.5. In overrun mode, the amount of fuel is reduced to very small values or to zero. Because the throttle valve position is not a direct measure of the amount of air, changes in air pressure and temperature are directly incorporated into the lambda value of the mixture supplied to the internal combustion engine. The pre-control values for the fuel quantity supplied to the internal combustion engine stored in the characteristic diagram must therefore be corrected by a superimposed control so that the lambda value can be set accordingly.

A particularly simple control method is a control for maximum power, which is only effective in the full load range, except in the idling range. The controller generates a factor by which the changes in the intake air quantity due to pressure or temperature fluctuations are taken into account. This factor, which is only determined in the full-load range, naturally also applies approximately to the map values of the partial-load range. For this reason, it makes sense to record this factor during the transition to the partial load range and to let it work in this area as well. Seen overall, this control factor influences the entire partial load and full load range, but is only determined when the internal combustion engine is operating at full load.

A block diagram of the control circuit is shown in FIG. Components identical to FIG. 3 are also identified by the same numbers. Only the deviation or innovation is described below. Since in the present case it is only a matter of regulating the maximum power, the test signal generator 26 acts via a summation point 80 and a multiplication point 81 solely on the injection times t read from the characteristic diagram 20 _; . Since two injection pulses are greased and leaned alternately, there is a speed-dependent influence. The controller 30, to which the output signals of the measuring device 27 are fed, has a multiplicative influence on the value read out from the characteristic diagram via the switch S2 in the case of full load. This controller works with a time constant that is as small as possible, with averaging at the same time using a mean value plot 82. When the full-load range is left, the controller 30 is switched off, the switch S2 is opened and the switch S1 is closed. Thus, the control factor stored by the mean value generator 82 becomes effective in the part-load range, specifically in such a way that the injection times t _; be influenced multiplicatively. At idle, there is also a regulation to the maximum power, so that controller 31 can also be used here in connection with switch S2. If the respectively required switchover or switch-off processes of the controller are solved by software, the area detection stage 83 has essentially symbolic meaning.

With the help of this arrangement, the characteristic map values of the injection time t _; Adapt to the changing operating conditions of the internal combustion engine for the partial load range by means of a full load control. These are particularly inexpensive and simple measures which can be implemented purely in software without great effort.

Situations can arise in which this recalibration after full-load operation is only relatively rare if the motor vehicle e.g. B. can be carried out for days only in the partial load range, for example in city traffic. Under certain circumstances, an infrequent recalibration of the map values leads to disadvantageous behavior of the internal combustion engine in the part-load range. A much more frequent redetermination of the control factor is achieved, however, if calibration can also be carried out in the partial load range.

To explain this method, FIG. 16 shows a section of the map in FIG. 12. Four map values, which are used particularly frequently in the partial load area, were selected for recalibration of the system. The mean value (t _i = 2.9 msec. For n = 1 200 and a = 7 °) is used in the normal part-load range. The upper value (t _; = 3.5 msec.) Corresponds to the injection time for this operating point of the internal combustion engine if the maximum output were controlled. This value was previously determined experimentally. If one of these four partial load points, which are shown in FIG. 16 as an example, is approached while the system is to be recalibrated, a change in the injection quantity of t in this example takes place for the duration of the calibration process _; = 2.9 msec. on t _; = 3.5 msec. instead of. The control for maximum output detects whether this preselected injection quantity corresponds to the maximum output in this operating range of the internal combustion engine. If there is a deviation that can be attributed to changed air temperatures or pressures compared to the normal state, a factor is determined which is this Change does justice. This factor is applied to the characteristic map values t in accordance with the previously described method _; attached for the partial load range.

In order not to unsettle the driver of the motor vehicle equipped with such an internal combustion engine due to the increased power output of the internal combustion engine during the calibration process, the power output of the internal combustion engine should not change due to the calibration process.

For this purpose, one can intervene in the ignition, namely that the increase in power of the internal combustion engine which is inevitably achieved by the regulation of the maximum power is compensated for by a reduction in the ignition timing. Once the control factor has been determined, the normal ignition timing can again be used together with the dimensioned map values using a new control factor.

On the other hand, it is also possible to change the increased power output of the internal combustion engine during a calibration process by determining the control factor only on a part of the cylinders of the internal combustion engine. However, separately controllable injection valves are a prerequisite for this. Some of the cylinders are switched to a maximum power control as described, but for the remaining cylinders the injection time is reduced to such an extent that the total power remains constant on average. In the present example (a = 7 °, n = 1 200), for example, half of the cylinders with the injection time t _; = 3.6 msec. and the other half of the cylinders with t _; = 2.3 msec. controlled. The determined control factor can now be applied to all cylinders. However, it can also be expedient to repeat the process with the remaining cylinders and then to use a control factor averaged over the cylinders.

An exemplary embodiment of this control method is shown in FIG. 17, the blocks identical to FIG. 15 being designated by the same numbers and also not being explained in more detail. In the present case, the injection valves are divided into two groups 23 and 23 '. Accordingly, there are also two multiplication points 81 and 81 'for the map which is supplied to the two valve groups 23 and 23'. As already described, these multiplication points 81 and 81 'are controlled via the summation point 80 either by the controller 30 or by the mean value generator 82. During the calibration process, for example, the valve group 23 with the increased map values for full load and the valve group 23 'with the reduced map values are activated to maintain an overall constant performance. Should small changes in output nevertheless occur in the form of changes in speed, a control circuit 90, which is sensitive to changes in speed, can compensate for them. For this purpose, a switch S4 is closed, so that the control circuit 90 acts on the valve group 23 'via the switch S4 and the multiplication point 81'. When the calibration process has ended, this switch S4 is opened and a switch S3 is closed, so that the addition point 80 is conductively connected to the multiplication points 81 and 81 '.

The second possibility for keeping the power output of the internal combustion engine constant during a calibration process is indicated by means of an ignition system 91 connected to the characteristic diagram 20. In this case, it is not necessary to divide the valves of the internal combustion engine into valve groups, since the increased power output of the internal combustion engine is realized by increasing the injection time during the calibration process by reducing the ignition point in the ignition system 91. For this purpose, a value for the withdrawal of the ignition angle of the internal combustion engine can then be stored in the map instead of the reduced value for the injection time.

With the aid of the arrangement shown in FIG. 17, frequent recalibration of the control factor for the characteristic map values of the injection time in the part-load range is possible, and thus an improved operating behavior of the internal combustion engine, in particular in the part-load range, is also guaranteed.

The invention is not restricted to fuel metering systems in which an intermittent injection, that is to say metering, is carried out over the opening period of the injection valves. It can also be used in an equally advantageous manner on electronically controlled injection systems with a continuous injection, such as is implemented in K or KE Jetronic systems, for example. The fuel is injected via a flow divider and the corresponding injection valves. The control piston of the flow divider is adjusted by electrohydraulic pressure actuators known per se. The pressure actuator is controlled by an electronic control unit, the main control variables of which are given by speed and load information (air mass, air volume, intake manifold pressure, throttle valve position). The use of a coarse, but very simple pilot control has been found to be particularly advantageous. B. a throttle valve speed pre-control by means of a map with a fine adjustment by a superimposed control. It is understood that the absolute values of the characteristic map sizes for a continuous injection must differ from those for intermittent injection, since either the injection quantities per stroke or per time unit are used as a basis. Furthermore, the following functions influencing the mixture metering must be fulfilled by the control unit: acceleration enrichment, full load enrichment, partial load reduction, lambda control, height correction reduction. In principle, the control methods already described, for example lambda controls or extreme value controls for regulating, for example, the consumption minimum, the maximum output or the smoothness of operation, come into consideration as controls superimposed on the pilot control.

Furthermore, the subject of the invention can also be applied to self-igniting internal combustion engines apply. The speed input and the accelerator pedal position instead of the throttle valve position can then be used as the characteristic input variable. The described exemplary embodiments can easily be transferred to self-igniting internal combustion engines by a person skilled in the field of mixture metering for internal combustion engines.

Claims (5)

1. Metering system for the fuel-air mixture in an internal-combustion engine with a control to a minimum specific fuel consumption which intervenes in the fuel metering by means of a test signal generator and in which the speed and a fuel metering signal are used as actual information for controlling to the minimum specific consumption value (be_min actual value), characterized in that the quotient of change in speed An and change in injection time Δte is determined forthe actual value detection, and in that the actual value is compared with a stored set point value, which is produced as slope of the torque curve over the injection time (AM/Ate) for the individual operating point at consumption minimum.

2. System according to Claim 1, characterized in that, for an internal-combustion engine with injection, the injection period is used as fuel metering signal.

3. System according to one of Claims 1 and 2, characterized in that the be_min control set point value is stored in a memory as a function of characteristic variables of the internal-combustion engine.

4. System according to at least one of Claims 1 to 3, characterized in that the set point values are preset as a function of the selected gear of the transmission connected to the internal-combustion engine.

5. System according to at least one of the preceding claims, characterized by its use in a self-igniting or externally auto-ignited internal-combustion engine with intermittent or continuous injection.

Images