EP1076764B1 - Method for automatically generating smoothed characteristic diagrams for an electronic engine control of an internal combustion piston engine - Google Patents

Abstract

The invention relates to a method for automatically generating smoothed characteristic diagrams for electronic engine controls of internal combustion piston engines. The invention is characterized in that the adjustment variable combination of the individual successive operating points are input by means of a motor control to a reference internal combustion piston engine by entering specified values of the boundary conditions for the operation of an internal combustion piston engine. According to the invention, the reference internal combustion piston engine is operated in this operating point, and the actual values and/or boundary conditions occurring during the same are acquired and compared with the specified values of the boundary conditions in an optimization system assigned to the engine control and, in the instance of variations, the adjustment variable combinations are optimally altered in a progressive manner by the optimization system. A quality function for the respective alteration of the adjustment variable combination is predetermined in the optimization system, and the quality function is corrected while taking into account already established values of the adjustment variable combination of at least one adjacent operating point.

Description

The invention relates to a method for automatic creation of smoothed maps for electronic engine control a piston internal combustion engine. Such procedures are known on the side (see WO-A-95 33 132).

In modern industrial society, mobility plays for him Transport of goods and great for trips to work Role. Much of this movement takes place on the street instead, and the piston engine plays as the drive source the dominant role.

Recently, the emissions from piston internal combustion engines have been has become the focus of public discussion. This is reflected in the legislation in the form of always lower emission limit values. Furthermore go up the prices for the required fuel. Both lead to that lower-emission and lower-consumption piston internal combustion engines required are.

To achieve this goal, piston internal combustion engines are developed and constructed according to the latest knowledge. It is not just a modern mechanical construction that plays a role a role, but the electronics comes through the enormously increasing possibilities and the flexibility, one ever increasing importance.

Where mechanical centrifugal force adjusters used to be the ignition point have adapted to the requirements is now an electronic Control unit in use. This can be a significant factor take closer account and more easily to different Purposes are adapted.

In these control units, the dependencies between input variables, for example speed, and the output variables, d. H. the adjustment variables, such as the ignition angle and injection quantity etc., stored in maps, for each operating state corresponding to a piston internal combustion engine Map points contain the current values for the calibration variables pretend.

When developing a piston internal combustion engine, the necessary maps are filled with values. So far the maps from particularly experienced developers of test bench measurements, using heuristic methods and Intuitively based on measurements on a reference machine created. This took considerable development time Claim and usually did not give optimal results.

The effort for the coordination of the maps depends heavily on the number of parameters to be calibrated. The takes Number of degrees of freedom in control units, for example through the introduction of exhaust gas recirculation (EGR), camshaft adjustment, variable intake system, to name just a few. The solution then required of a more than three-dimensional one Optimization task with many parameters is hardly for people still manageable.

For this reason, systems for automatic map optimization and developed appropriate software. Create this Maps based on test bench measurements and mathematically sound algorithms. So there are fewer road tests with vehicles required, and optimization of Piston engine is possible, even if the entire vehicle does not yet exist. On the one hand, this means the development time and thus the "time to market" is shortened and consequently there is a cost saving. On the other hand, they are generated Results reproducible and not human Optimizer dependent who works with intuition. The optimization system is also easier to adapt and to others Adapt specifications.

Because of the relatively short time required, the automatic Optimization carried out several times with different configurations become. This opens up various possibilities Play through scenarios in a practical experiment with reasonable Effort would not be feasible.

With the methods used so far, it is possible for a given construction of a piston internal combustion engine To create "mother" maps, after which for the later Series production and also for the engine control to be manufactured in series the corresponding map data medium are created can. The disadvantage of the previously used method is however, in that while the automatic Optimization for each support point or for each operating point of a map, a value is generated, but without the Connections between neighboring support points must be observed. This leads to jumps in the calibration data of neighboring ones Support points that determine the transferability of the optimization result and endanger driveability in practical vehicle use. Strong jumps in calibration data from neighboring operating points must therefore be avoided.

Jumps occur in two phases of optimization: One is the problem that voting results are within a map area optimized according to the same criteria have such jumps in adjustment. Second results another problem of erratic transitions when merging of optimized according to different criteria Map areas.

The invention is based on the object of a method find that avoidance already during the optimization run too strong jumps in the calibration data and nevertheless allows a good optimization result and the creation of a smoothed map allows.

This object is achieved with the method steps specified in claim 1 solved. Inventive modifications of the process are given in claims 2 to 4.

Fig. 1

ein Blockschaltbild für einen Prüfstand mit Kennfeldoptimierung,

Fig. 2

den Arbeitsablauf des Prüfstands gemäß Fig. 1 als Blockschaltbild,

Fig. 3

ein ungeglättetes Kennfeld, erstellt nach dem Verfahren gemäß dem Stand der Technik,

Fig. 4

ein geglättetes Kennfeld, erstellt nach dem erfindungsgemäßen Verfahren,

Fig. 5

die Darstellung eines Verstellgrößensprungs für eine variable Stellgröße,

Fig. 6

die Darstellung des Verstellgrößensprungs gemäß Fig. 5 in einem Koordinationssystem für zwei Variable,

Fig. 7

ein Flußdiagramm für eine Kennfeldoptimierung mittels einer vorgegebenen Gütefunktion,

Fig. 8

ein Detail-Flußdiagramm zur Erläuterung der Optimierung der Ziel- und Grenzwertgrößen,

Fig. 9

die Wirkungsweise einer Überlagerung einer Gütefunktion mit einer Incentivfunktion,

Fig. 10

ein Detail-Flußdiagramm für eine Kennfeld-optimierung bei Überlagerung einer Gütefunktion mit einer Incentivfunktion,

Fig. 11

ein Detail-Flußdiagramm für eine Kennfeld-optimierung mittels Gütefunktion und Verstellgrößendifferenzerfassung,

Fig. 12

ein Detail-Flußdiagramm für eine Kennfeldoptimierung zur Begrenzung der Rauhheit in jeder Betriebsstufe.

The invention is explained in more detail below with the aid of schematic drawings. Show it:

Fig. 1

a block diagram for a test bench with map optimization,

Fig. 2

1 as a block diagram,

Fig. 3

an unsmoothed map, created according to the method according to the prior art,

Fig. 4

a smoothed map, created according to the inventive method,

Fig. 5

the representation of an adjustment variable jump for a variable manipulated variable,

Fig. 6

5 in a coordination system for two variables,

Fig. 7

1 shows a flow diagram for a map optimization by means of a predetermined quality function,

Fig. 8

a detailed flow chart to explain the optimization of the target and limit values,

Fig. 9

the mode of action of a superposition of a quality function with an incentive function,

Fig. 10

1 shows a detailed flow chart for a map optimization when a quality function is overlaid with an incentive function,

Fig. 11

a detailed flowchart for a map optimization by means of quality function and adjustment variable difference detection,

Fig. 12

a detailed flow chart for a map optimization to limit the roughness in each operating level.

Fig. 1 shows a test rig with automatically-working map optimization system 1, to input information I _and output information I _from as well as a map output K _from, electric motor control device 2, reference piston-type internal combustion engine 3 for a series of the required measuring devices and 4. The input information of the system part predefined by the user (limit values, targets and map points to be optimized) and partly requested by the system during the optimization of the engine test bench (measured values). For this purpose, the system specifies adjustment variables that are automatically set on the piston internal combustion engine 3 and then evaluates the measured values to determine optimal adjustment variables. Finally, the system generates maps as a result, which are transferred to the engine control unit 2 of the piston internal combustion engine 3 and for which the optimization was carried out. The engine control unit 2 also takes into account all the values that are relevant for the use of the piston internal combustion engine 3 in a given vehicle.

In Fig. 2 the workflow of the test bench from Fig. 1 is with exemplary input information and examples for calibration variables reproduced, for each to create a map and which measured values can be recorded here. The individual components of the test bench are here with the reference symbol identified from Fig. 1. Both for the engine control unit 2 of the test bench and for the measuring device 4 is indicated that further control elements and measuring devices can be provided. The map optimization system determines during the automatic map optimization for each map point (Map point = a combination of the input variables), for example load and speed an adjustment value, for example the ignition timing. However no relationships between neighboring map points respected.

As shown in FIG. 3, this type of map generation results Jumps in the calibration variable values to neighboring ones map points that show the transferability of the optimization result into the engine control unit and the driveability in practical terms Endanger vehicle use. Big leaps in adjustment size Neighboring map points must therefore be avoided. It a "smoothed" map must be generated, like this for Comparison in Fig. 4 is. A smooth map is characterized by small jumps in the adjustment size.

The change in the adjustment variable, which is used to evaluate the smoothness of a characteristic diagram, is explained with reference to FIG. 5. For the sake of clarity, an example is shown in which only one adjustment variable, here the ignition timing ZZP, is considered.-The ignition timing in this example depends only on a variable input variable, here the speed n, while the value for the torque remains constant is held. Shown are a speed n _a , called "current speed" and two neighbors "n1" and "n2". The current speed has the ignition timing ZZP _a and the two neighbors have the ignition timing ZZP1 and ZZP2.

At the current engine speed n _a , an "ideally smooth ignition timing" is determined, which leads to a smooth map. To determine this "ideal ignition timing" at the current speed, an interpolation between the ignition timing of the neighbors is carried out, shown in FIG. 5 by a dashed line between ZZP1 and ZZP2. The difference between this straight line and the ignition timing ZZP _a at the current speed is defined as an adjustment variable jump. The smaller the adjustment variable jump (here the ignition timing jump), the smoother the map at the current point (here at the current speed) in relation to its neighbors.

For calibration variables that usually change linearly, this happens the determination of the ideal variable value by linear interpolation. Generally speaking, however other interpolations are used.

The map points normally depend on (at least) two Input variables, for example the ignition timing of engine speed n and load M. In this case there are more than two neighboring ones Map points between which the ideal calibration variable value must be interpolated, as Fig. 6 shows. The representation according to FIG. 5 is drawn into the coordinate system of FIG. 6. To get a smooth map, it is enough does not perform the interpolation specified in FIG. 5, but also the values of the other neighboring ones must be added Map points, for example N7 and N3, are taken into account.

The same procedure is followed for other adjustment variables, e.g. Injection quantity, start of injection, exhaust gas recirculation rate etc. In In these cases there is an interpolation for each calibration variable between the neighboring map points to determine the ideal adjustment variable value carried out.

A so-called quality function is used to determine the cheapest combination of adjustment variables. The optimization goal is to fall below the specified limit values (e.g. for exhaust gas emissions). The quality function is made up of all the variables G ₁ to G _n to be optimized (e.g. consumption, emissions, ...) and the associated limit values GW ₁ to GW _n . The weight of the individual quantities in the quality function is determined by factors λ ₁ to λ _n . Thus the quality function is: Goodness = λ 1 (G 1 - GW 1 ) + λ 2 (G 2 - GW 2 ) + λ 3 (G 3 - GW 3 ) + ... + λ 3 (G 3 - GW n )

As an example, a quality function for an optimization of the fuel consumption b _e with a simultaneous requirement for compliance with a nitrogen oxide limit value (NO _x ) is given.

When NO _x the current measured NO _x value designated and NO _xMax to be observed threshold and b _e the current measured fuel consumption, so the quality function is for this application: quality E.g = λ 1 (NO x - NO Max ) + λ 2 * b e

A minimum of the quality function is determined during the optimization. The sequence of such an optimization in the map optimization system 4 is explained and illustrated in FIG. 7 in the form of a flow chart. In the example mentioned, the ZZP is varied until the minimum of the quality function is found. If the limit value for NO _{x is} still exceeded at this minimum, the quality function can be trimmed to a greater sensitivity to the nitrogen oxide value by varying the Lagrangian factors λ ₁ and λ ₂ and a minimum can be sought again.

The variables to be optimized are a function of the calibration variables and the map point: G n = f (calibration variables, input variables)

For the example mentioned, this means: NO x = f 1 (ZZP, n, M) and b e = f 2 (ZZP, n, M)

The minimum of the quality function for the entire map is determined by the minimum of the quality function in each map point is determined by varying the adjustment variables, as in Fig. 8 shown. The following applies to the selected embodiment for a map point that n and M are kept constant and the minimum of the ZZP is determined. The determination of the minima is carried out in every map point. The adjustment values, that belong to these minima are the optimal adjustment values regarding the optimization goals in each Map point. The result of this procedure is an unsmoothed one 3, which is still significant Adjustment size jumps.

In order to avoid jumps in the adjustment size, the Calculation process to optimize the quality function become. This will cause map jumps to occur in the Avoid run of optimization. The smoothness of the to be created For this purpose, the map is used as an additional boundary condition taken into account in the optimization.

In a first embodiment of the method according to the invention This is done in that an adjustment variable combination in the calculation process, which leads to a smooth map, is "rewarded", so that it is preferred over others in the optimization Adjustment variable combinations, the same or even better results deliver with respect to the other boundary conditions, but too lead to larger jumps in the adjustment size.

The quality function is influenced by a so-called incentive function for rewarding favorable adjustment variable combinations with regard to smoothness, which can be formulated as follows: Incentive = | a (VG1 - Opt1) | + | b (VG2 - Opt2) | + | c (VG3 - Opt3) | + ... + | d (VGx - Optx) |

VG1 to VGx denote the calibration variables, Opt1 to Optx the optima of the corresponding calibration variables in the neighboring ones Operational stages. a to d are factors that influence the influence of Determine the respective adjustment variable in the incentive function.

As an example, the incentive function for the ignition timing ZZP is shown as an adjustment variable, M1 being the optimum of the ignition timing from the neighboring operating stages. The optimum is the "ideal variable value", ie the interpolated value from the optima of the neighboring operating levels: incentive E.g = | a (ZZP-M1) |

This incentive function is superimposed on the quality function. The result is a new quality function that has a different minimum and thus leads to a different combination of adjustment variables: quality incentive = Goodness + incentive

For the example, the overlaid function is: quality Incentive Ex = | λ 2 (NO x - NO xMax ) + λ 2 * b e | + | a (ZZP - M1) |

The mode of operation of such an incentive function is shown in FIG. 9 played. The ignition timing should be optimized (variable) taking into account minimal consumption (Target). In this case, the quality function is the course of consumption above the ignition point. It is said to be smooth Transitions to neighboring map points are generated.

At a map point "a", the ignition point x was considered optimal determined in terms of consumption (Fig. 9). In the neighboring Map point 2 should now optimize the ignition timing taking smoothness into account. In this map point "b", the ignition point y would be optimally determined in terms of consumption because the minimum M2 is smaller than the minimum M1 (Fig. 9).

However, the adjustment variable combination in the minimum M1 leads to one greater smoothness than the adjustment size combination in the minimum M2, as the optimal one for the neighboring map point 1 Adjustment variable combination for the minimum M1 and not for the minimum M2 lies.

To influence the smoothness, the quality function quality _{example is added to} the incentive function incentive _example , which has its minimum at the ignition point x of the map point "a", the function value of which becomes less favorable the further the ignition point differs from the ignition point x (Fig. 9). The addition results in the new quality function, quality _{incentive example,} for the map point "b" (FIG. 9). When optimizing in map point "b", the ignition point is now found in the minimum M1, which is closer to the ignition point x of the adjacent map point than the ignition point y. This leads to a more favorable adjustment size combination with regard to smoothness.

This procedure is now applied iteratively to all map points. Each map point is thereby in the optimization runs alternately both neighbor, the influence on the straight point to be optimized, as well as to be optimized Point affected by its neighbors. In general In case of multiple neighbors, an incentive function is used the correspondingly several minima depending on the optimal ones variable of the neighbors. The example uses one linear incentive function. Depending on the course of the quality function and other sizes involved, however, depending on Requirement to use non-linear incentive functions come to the described influence on the quality function to reach.

The iterative process of a map optimization with influencing is explained by an incentive function in FIG. 10 and shown.

In another embodiment of the method, the adjustment variable jump a map point a measure of the smoothness determined on this point.

For this purpose, the difference between the ideal value and the value found during optimization. This difference is called the adjustment variable difference. The adjustment variable difference is, like other boundary conditions, z. B. the emission values, included in the optimization.

The adjustment variable difference is included in the optimization as an additional constraint instead of the incentive function. For this purpose, it is treated like a measured value of the piston internal combustion engine. With each measurement on the piston internal combustion engine, it is calculated from the calibration variables of the neighboring and the current operating stage. The adjustment variable difference, like the exhaust gas emissions, is included in the quality function. So one of the values G ₁ to G _n can contain the smoothness information: Goodness = λ 1 (G 1 - GW 1 ) + λ 2 (G 2 - GW 2 ) + λ 3 (G 3 - GW 3 ) + ... + λ 3 (G n - GW n )

For an optimization of fuel consumption and nitrogen oxide development with specification of a maximum roughness R _max (roughness = opposite of smoothness) one obtains: quality VdvBsp = λ 1 * b e + λ 2 (NO x - NO xMax ) + λ 3 (R - R Max ) if R is the currently determined value for the roughness.

The procedure for an adjustment variable is described below. If there are several calibration variables, the procedure is applied for each variable.

One considers an operating level in the map, called the current one Operation level BS, and their neighbors. During optimization this operating level are the calibration variable values of the neighbors constant because only the calibration variable value of the current operating level is varied. From the calibration variable values of the neighbors becomes the optimal calibration variable value for the current operating level calculated.

At the current operating level there is a minimum of the quality function searched. The adjustment variable of the current point becomes this varies to find the minimum as shown in the flow chart Fig. 11 can be seen. The result is for each variable value another adjustment variable difference corresponding to the variable smoothness of the variable to the neighbors.

At the end of an optimization cycle (optimization of all map points), a global value R is calculated for the roughness. "Global" means: for the entire map. To do this, all differences in the calibration variables are added up. This roughness value is compared with the global limit value for the roughness R _max . A small limit value corresponds to a small roughness corresponding to a good smoothness of the map.

If this limit value is exceeded, the factor λ (λ ₃ in the example above) of the roughness in the quality function is modified, preferably increased, such that the roughness has a stronger influence on the quality function. In the next optimization run, the optimal calibration variables for the changed quality function are determined. Since this quality function is more dependent on the roughness, more favorable values for the adjustment variables with regard to smoothness are achieved.

By specifying a global limit, the roughness limited for the entire map. It doesn't matter what share the individual operating levels in the overall result have, but only that the limit is not reached. This process is repeated until all optimization goals can be achieved.

When the roughness is limited by a global limit local, existing adjustment size differences due to smooth Parts of the map are balanced in the total value of the roughness become. "Local" means: In a map point. Local roughness are undesirable, however.

In order to keep these local adjustment variable differences small, is the roughness of the map in each individual operating level by introducing and specifying a local limit R (n, M) limited. This leads to combinations of variable sizes, that exceed this limit, immediately during optimization this map point are rejected, as in Fig. 12 indicated.

The maps of piston internal combustion engines are divided into several Areas divided in which different boundary conditions and Optimization goals apply. An area is legal through that prescribed driving cycle (to limit emissions) is specified and is called the driving cycle area. other areas are the full load curve on which maximum performance is required and the rest of the map, in which is usually minimal Consumption is desired, called the consumption minimum area.

To make a statement about the roughness in the entire map the roughness values of the different areas summarize accordingly. This is done using the following procedure applied:

After the optimization is complete, there is a value for each area for the roughness. The optimization system calculates this Value for each area based on the length of stay from the results of the individual operating levels corresponding to Consumption and emissions.

Dwell times are only for the driving cycle area through the driving cycle specified. The number of operating levels and the length of stay in the individual operating levels (for the driving cycle area) are converted into stationary by converting the driving cycle Operation levels determined. For the full load curve and the consumption minimum area there are no corresponding requirements.

To optimize on the full load curve and in the consumption minimum area to be able to carry out, however, will also remain there needed. In principle, any length of stay be accepted. However, since the dwell times are also used for the projection The following procedure is used for the roughness to determine the dwell times for full load curve and minimum consumption area applied:

The average length of stay in an operating level for the driving cycle area can be calculated from the length of stay and the number of operating levels in the driving cycle area: Average length of stay = Seconds in driving cycle area / number of operating levels in Driving Cycle area

This average length of stay is also used for the operating levels on the full load curve and in the consumption minimum area used. This is a calculation of the roughness for that Entire map possible: The results of all operating levels are extrapolated (on average) with the same length of stay. The share of an area in the overall result is obtained therefore as a ratio of the number of operating levels in the area to the total number of operating levels in the map.

With the method shown, a smoothed map can be made generate, as can be seen from the comparison between FIGS. 3 and 4 leaves. This smoothed map not only enables the fulfillment of emission limit values, like the map according to Fig. 3, but by the smooth transitions between the operating levels are the transferability to the engine control unit and the driveability ensured. The so when operating a reference piston internal combustion engine created, smoothed maps then serve as "mother" maps for the production of Engine control units for piston internal combustion engines of this type.

Claims (4)

1. A method for automatically generating smoothed characteristic diagrams for electronic engine controls on piston-type internal combustion engines, characterized in that on a reference piston-type internal combustion engine by means of an engine control the adjustment variable combination of the individual successive operating points is input by entering desired values for the marginal conditions for the operation of a piston-type internal combustion engine, that the reference piston-type internal combustion engine is driven at this operating point and the resulting actual values and/or marginal conditions are detected and, in an optimization system which is assigned to the engine control, are compared to the desired values for the marginal conditions and that in case of deviations, the adjustment variable combinations are changed and optimized step-by-step with the optimization system, whereby a quality function for the respective change in the adjustment variable combination is predetermined in the optimization system, and that the quality function is corrected, respectively by taking into account already fixed values for the adjustment variable combination of at least one neighbouring operating point.
2. A method according to claim 1, characterized in that the quality of the respective adjustment variable combination is defined by the function Quality = λ1(G1 - GW1) + λ2(G2 - GW2) + λ3(G3 - GW3) + ... + λn(Gn - GWn) λi = factors, Gi = variables to be optimized and their asociated limit values GWi.
3. A method according to claims 1 and 2, characterized in that by taking into account fixed characteristic diagram values of at least one of the neighbouring operating points, an incentive function is superimposed on the quality function: a, b, c... = factors, Incentive = |a(VG1 - Opt1)| + |b(VG2 - Opt2)| + |c(VG3 - Opt3)| + ... + |d(VGx - Optx)| VGi = adjustment variables i and their associated optima VG; i = 1...X
4. A method according to claims 1 and 2, characterized in that a maximum permissible roughness for the characteristic diagram to be generated is specified for the respective adjustment variable and is taken into account for the quality function.

Images