Method for operating an internal combustion engine installed in a vehicle

Abstract

The invention relates to a method for operating an internal combustion engine installed in a vehicle, in particular a diesel engine, in which the instantaneous concentration of a pollutant contained in the exhaust gas, in particular the NO.sub.x concentration in the exhaust gas, is measured or calculated in the flow direction after an exhaust gas aftertreatment. Using the determined pollutant concentration, the predefined distance- and/or power-based compliance with pollutant limiting values in mg/km or mg/kWh are monitored by means of specifically influencing the operating parameters of the internal combustion engine and/or an exhaust gas after treatment system in regulated form.

Claims

1. A method for regulating an internal combustion engine installed in a vehicle, wherein the method comprises the steps of evaluating with an engine control system an instantaneous pollutant concentration contained in exhaust gas, in the flow direction after an exhaust gas aftertreatment and, monitoring a predefined distance of travel based compliance with pollutant limiting values in mg/km, and specifically regulating operating parameters of the internal combustion engine or an exhaust gas aftertreatment system based on the instantaneous pollutant concentration so that a limiting value is cumulatively adhered to across a specified minimum distance of travel.

2. The method as claimed in claim 1, wherein in addition to the instantaneous pollutant concentration, also at least one of an instantaneous fuel consumption or a foreseeable fuel consumption corresponding to a respective driving situation, depending on a respective driving style, a respective vehicle, vehicle status and environmental factors are taken into consideration, in the case of regulated operation of the internal combustion engine.

3. The method as claimed in claim 1, wherein an instantaneous status of an aging process of the exhaust gas aftertreatment system, fuel quality or the quantity, type or quality of chemical compound added to the exhaust gas aftertreatment for a pollutant reduction is/are taken into consideration during regulation.

4. The method as claimed in claim 1, including the step of regulating manipulation of injected fuel quantity, at a point in time of injection, type of the fuel injection, boost pressure of added combustion air, temperature of the combustion air, temperature or quantity of recirculated exhaust gas, valve control times or engine compression ratio.

5. The method as claimed in claim 1, wherein during regulation, a pollutant limiting value allocated to a respective time or respective vehicle position is taken into consideration.

6. The method as claimed in claim 1, wherein, during regulation, engine characteristic maps stored in the engine control system, which take different variants into account for the operation of the internal combustion engine, are used.

7. The method as claimed in claim 6, wherein during regulation, an interpolation is carried out if no engine characteristic map corresponding to a respective operational state of the internal combustion engine is available.

8. The method according to claim 6, wherein a selection of the different variants is carried out during regulation taking control deviation into consideration without interpolation between the different variants, a regulation depending on the control deviation with corresponding interpolation or conversion of a reference variable into an operating-point dependent target value for raw emission of the pollutant in g/(kW.Math.h) is carried out and a regulation is carried out as claimed in a modification depending on the control deviation.

9. The method as claimed in claim 1, wherein a regulation takes place so that a predefined pollutant limiting value of a distance of at least 2 km is adhered to.

Description

(1) The figures show:

(2) FIG. 1 a schematic illustration for the regulated adaptation of the engine control system for complying with the required nitrogen oxide limiting value during real driving operation;

(3) FIG. 2 a comparison of measurement and simulation for the specific CO.sub.2 and NO.sub.x emission in the case of various driving cycles;

(4) FIG. 3 schematically, the integration of the MATLAB/SIMULINK model for controlling the motor into a 1D simulation model;

(5) FIG. 4 a coupled model for calculation of driving cycles (vehicle model engine operating model application model) and

(6) FIG. 5 a diagram of emissions across a specifiable cycle time.

(7) The basis for a developed model to calculate the engine process as well as the executed tests formed a four-cylinder diesel engine with a common-rail direct injection system with exhaust gas turbocharging. The 2.0 l series engine reaches a maximum full-load torque of 330 nm and provides up to 110 kW at the rated output point. Thereby, as a standard, a water/air-charge-air cooler is used for cooling the compressed intake air, which is operated in a separate low temperature circuit. The engine internal reduction of NO.sub.x emission takes place via a combined EGR system, which consists of low-pressure and high-pressure exhaust gas recirculation (LP- and HP-EGR). In the case of the LP-EGR, the exhaust-air partial flow removed after the DPF is cooled before introduction into the compressor line by means of a heat exchanger. In addition, a flap gate in the exhaust system partially implements the necessary pressure gradient for this. The high-pressure EGR is carried out uncooled and, thereby, the recirculated exhaust gas is introduced into the cylinder after the charge-air intercooler and before the inlet channels. For the engine external reduction of pollutant emission (CO, HC, soot), the diesel oxidation catalytic converter (DOC) and the subsequent DPF are arranged close to the motor. By means of the aforementioned exhaust gas aftertreatment system, the engine fulfills the Euro-5 emissions standard.

(8) The developed simulation model in the simulation environment GT-SUITE/GT-POWER (Version 7.4.0, Build 4) comprises all relevant components of the internal combustion engine. Starting from the realistic representation of the charge exchange and the modeling of the various wall heat transfers and heat capacities, the combustion process and the raw exhaust gas emission are predictively calculated with the aid of phenomenological models. The required calibration and validation took place based on available measurement data from the highly dynamic engine dynamometer. From the viewpoint of the general pre-calculation capacity and qualitative assessment, in this way, a sufficiently good concordance can be achieved.

(9) A vehicle model was used for the illustration and the simulation of driving cycles. The model developed in MATLAB/SIMULINK for highly dynamic engine dynamometers was coupled to the engine model of the 1D simulation via appropriate interfaces. All relevant characteristics of the vehicle and the drivetrain (e.g. gearing and clutch) as well as the boundary conditions occurring during the real ride depicted the vehicle model. An integrated driver model assumes the task of speed control. Depending on the speed difference (v.sub.current−v.sub.target), the calculated torque requirement is passed on to the engine operating model. With the aid of the recirculation value, the torque calculated within the engine operating model, the vehicle operating model determines the speed, as well as the corresponding engine speed. The latter is also specified for the engine operational model.

(10) In FIG. 2, the distance-specific emission of CO.sub.2 and NO.sub.x for a current middle-class vehicle is shown as an example for various driving cycles, thereby indicating a comparison between measurement and simulation for the specific CO.sub.2 and NO.sub.x emission during various driving cycles. At this point, it must be noted that the low existing deviations between measurement and simulation accumulate from the previously mentioned stationary validation for this comparison case, thereby being able to portray a very good result.

(11) The implementation of the aforementioned objective can be achieved by means of another control model, which is based on different variants of the engine application. For this purpose, applications with a low specific fuel consumption, meaning a high process efficiency, with a low NO.sub.x emission, as well as intermediate variants should be created.

(12) The following explanations should demonstrate a possible application method and exemplary manner and should by no means be deemed complete. The term “variant” is used in the following for the respective version of the engine application and is specified with V 0.0 to V 1.0. Thereby, the design, implementation, and testing is carried out only by using an engine process simulation.

(13) The application V 0.3 forms the starting point of variant development and essentially corresponds to the application of the underlying basic engine. A consumption-specific best variant (V 0.0) was generated, having been derived from this, which is, among other things, optimized with regard to the center of combustion and EGR rate within permissible limits (air ratio, cylinder pressure increase, etc.) For the motorized operation with minimum NO.sub.x emission, the EGR rate should be increased to the extent that the desired NO.sub.x emission has been configured. At the same time, an increase in charge pressure should take place in order to keep the combustion air ratio constant in comparison with the basic variant. If a target variable exceeds the specified limiting value, the torque is reduced to the last depictable operating point of the diesel engine. Furthermore, the isolines (lines with a constant EGR rate) in the generated EGR characteristic map run parallel to the effective intermediate pressure in contrary to the basis. By means of this, the nitrogen oxide raw emissions can be kept almost independent from the engine speed. In order to improve the regulation quality of the model, furthermore, an application V 0.7 with interpolated EGR rates between the basic variant (V 0.3) and the variant with a minimal NO.sub.x emission (V 1.0) can be formed. The design of the distribution between LP- and HP-EGR can take place with regard to the optimal efficiency range for the compressor and turbine.

(14) Variants of the engine operation with an identical division between LP and HP-EGR as well as the achievable full-charge curves, at a consumption optimal all the way to a minimal NO.sub.x emission, and the differences of the individual variants can be recognized at two different partial charge operating points. As the NO.sub.x emission decreases, the specific fuel consumption continually increases. At the same time, the air ratio (with limitations in the case of the consumption-optimum variant) remains constant.

(15) In addition to the emission differences, the various tested variants have considerable differences with regard to dynamic behavior. In particular, the regulation at high EGR rates in the range of full charge delays the building up of boost pressure, thereby increasing the time until the desired torque has been achieved. In addition, the specification of the EGR rate during genset loading plays a crucial load. The rate reduction with an existing boost pressure difference leads to an improved dynamic behavior and a lower soot emission, however at the same time, it causes a higher NO.sub.x emission. Based on an elasticity test during acceleration of the vehicle, differences also occur due to the influence of the EGR take-back (EGR reduction) while the genset loading as well as the shifting strategy (used with and without shifting back). Higher EGR target values increasingly increase the time until the target speed is reached. At the same time, a considerable decrease in NO.sub.x emission with a significant increase of soot emission associated therewith can be recorded. Doing away with the reduction in EGR intensifies this effect. The temporal delay can be reduced by shifting back to greater gear ratios, wherein the NO.sub.x emission decreases due to the lower engine torque, however, the particle and CO.sub.2 emission increase in the process.

(16) The created simulation model, which has already been addressed, obtains different engine characteristic maps for the control and regulation of the respective process variable. Via the target values, engine speed and engine load, reference and pre-control variables, such as boost pressure, EGR, position of the boost pressure regulator/VTG setting, fuel mass and points in time of injection are stored. For the simultaneous provision of the different application variants, the mentioned engine characteristic maps are transferred into a MATLAB/SIMULINK model. Via a corresponding interface, only the target values can be provided to the 1D simulation program, which correspond to the selected variant. Furthermore, the model calculates the distance-specific emissions and makes the implementation for the control and variation of the variants possible.

(17) FIG. 3 schematically shows the integration of the MATLAB/SIMULINK model for controlling the engine into the 1D simulation model, the coupling to the vehicle model as well as the transfer process and control variables.

(18) The various generated variants are, as has been explained previously, stored in diverse engine characteristic maps in the MATLAB/SIMULINK model. According to the selected variant as well as the target values for engine speed and load, the process control variables (e.g. EGR rate) can be taken from the respective engine characteristic maps and passed on to the engine operating model. For the case that no engine characteristic maps exist for the specified variant, the interpolation between the values of contiguous engine characteristic maps is possible. FIG. 4 schematically illustrates this approach.

(19) The basis for the selection and control of the respective variants represents the continuous monitoring of exhaust gas emission. The calculated emission values from the engine operating model are recorded by the MATLAB/SIMULINK model, temporally integrated, referred to the distance driven and used for the engine operation as a control variable.

(20) The specification of the NO.sub.x limiting value to be satisfied in [g/km] is used as a reference variable for the variant control. For implementing the regulation and, if applicable, the control, three different regulation versions were analyzed in detail: #a Variant shifting depending on the control deviation (without interpolation between the variants) #b Controlling the variants depending on the control deviation with corresponding interpolation #c Conversion of the reference variable into an operating-point-dependent target value for the raw emission in [g/(kW.Math.h)] and corresponding modification depending on the control deviation.

(21) In the latter case, depending on a stored drive resistance curve, the reference variable is converted into a target value for raw nitrogen oxide emissions in [g/(kW.Math.h)]. Based on the comparison with the current raw nitrogen oxide emission calculated in advance, which can also be stored in corresponding characteristic maps for each variant in the simplest case, the selection of the required variant takes place. Under the assumption of a continuous transfer between the variants with regard to NO.sub.x emission, a variant interpolation, meaning an interpolation between the required engine characteristic maps, is permissible. In addition, a continuously linear controller, PID controller, modifies the reference variable, instantaneous NO.sub.x emission with the aid of a downstream offset in order to ensure that the reference variable and the target value are achieved and complied with.

(22) The control model in MATLAB/SIMULINK furthermore obtains a simplified calculation model to depict an exhaust gas aftertreatment system for the selective catalytic reduction by means of urea. This contains the characteristics for the light-off behavior (i.e. the progression/development of the conversion rate over the temperature, among other things), the minimum temperature as well as the dosage quantity ratios for the urea injection and the deterioration of the conversion behavior at excessively high spatial speeds. Using the calculated implementation efficiency (limited to maximum of 95%) the NO.sub.x emission is determined according to the SCR system, which, in turn, can be used as a new control variable. Particularly for the regulation version #c, the efficiency of the exhaust gas aftertreatment can thereby be taken into consideration and, accordingly, the target value specification leads to the increase of the raw nitrogen oxide emissions to be configured.

(23) The calculated reduction of nitrogen oxide cannot be taken into account by the calculation model in this case for the recirculated exhaust gas (LP-EGR) and the intensifying effect thereby achieved in the case of raw nitrogen oxide reduction.

(24) In the following, the results of the analyses will be presented for the variable application on a diesel engine. First of all, the certification-relevant test cycles are used for verification of functionality: New European Driving Cycle (NEDC), Worldwide harmonized Light vehicles Test Cycle (WLTC), Federal Test Procedure (FTP75), Aggressive Driving Cycle (US06) as well as the newly developed internal RDE cycle, which should fulfill the current legal requirements subject to complying with the prescribed boundary conditions. Following this, different variations show the universal applicability of the engine operation strategy developed.

(25) The function of the engine operating strategy can be demonstrated using the example of the NEDC using a current middle-class vehicle with a mass of approximately 1.7 t. Three versions of regulation with an emphasis on the results of the CO.sub.2, NO.sub.x and soot emission were analyzed. The consideration initially occurred without taking the SCR system into consideration. The distance-specific nitrogen oxide raw emission served as a control variable. Due to the formation of quotients for the control variable (emission mass NO.sub.x per distance unit), the regulation for a distance s.sub.min has been set with a value of 2 km. Variant V 0.3 serves as an initialization. In the case of the operating-point-dependent regulation version #c, the initialization is limited to a constant value of the downstream offset for the distance s.sub.min.

(26) With all three variants for the regulation, the defined NO.sub.x limiting value in g/km could be precisely adhered to in a sufficient manner. In comparison with the basis, however, with reference to the CO.sub.2 and soot emission for the versions #a and #b, greater differences were shown, the extent of which was considerably coupled to the settings of the respective controller. This behavior was shown in all examined driving cycles (WLTC, etc.). The background entails an only insufficiently possible selection of a generally valid setting of the PID controller. The result achieved is correlated with the dynamics of the controller and, at the same time, with the length of the text cycle. This results in a trade-off in controller configuration.

(27) The best results could be achieved using the control version #c. The speed-dependent variant control at first approximation requires a considerably lower control intervention and allows a comparably “slow” control parameter with a greater time constant. In contrast, for this version, the variant used changes almost for each time step. For limiting and stabilization, the regulation only occurred for speeds over 2 km/h and load conditions above 5 Nm. Otherwise, the variant to be used has been specified directly. The intervention of the controller for the offset (control behavior) was, in particular, shown in the idling phases of the vehicle due to the slow change of the respectively chosen variant.

(28) Furthermore, in the case of this variant of the regulation, the advantage arose that the effect of the SCR system was taken into consideration directly. Therefore, for further analyses, only this variant was further pursued.

(29) The function of the regulation taking the SCR system into account should be described in the following using the example of the future certification cycle WLTC with the same vehicle configuration. At the beginning of the certification cycle, the start temperature of the engine as well as all gas-bearing components was 25° C. For the function of the SCR system, in addition to the actual catalytic converter temperature also the gas temperature at the urea injector must be taken into account. In addition to the exhaust gas temperature, the implementation level achieved across the cycle time using an SCR system also plays a role. As a reference variable for the regulations, a NO.sub.x target limiting value of 80 mg/km has been defined and an initialization distance s.sub.min of 2 km has been defined.

(30) The light-off of the SCR system is achieved approx. 750 s after the cycle starts. As of this point in time, a considerable increase of nitrogen oxide raw emission is possible so that the engine can be operated with the consumption favorable variant. If the implementation level of the SCR system (e.g. approx. at second 1,500) drops on account of a high spatial speed in the catalytically coated sDPF due to a strong vehicle acceleration or an exhaust gas temperature that is too low, the variant is immediately changed, thereby being adapted to the instantaneous nitrogen oxide raw emission value.

(31) The main advantage of the presented engine operating strategy results for use during real driving operation. The selected RDE cycle for the functional verification was divided almost evenly within the so-called third matrix (city, suburban, and highway). In relation to the driving dynamics, the driving style was in the upper third of the permissible range (values for v.Math.a.sub.pos,>0.1).

(32) The settings of the controller for the selection of the variants, which resulted from the optimization during the certification cycles were taken on for the RDE ride carried out here. Since, for each partial section, the required NO.sub.x limiting values must be adhered to, these settings have also been kept constant during the course of the cycle.

(33) Initially, with no regard to the SCR system, regulation was performed to meet the nitrogen oxide value of a basis and then to a NO.sub.x limiting value of 80 mg/km with an additional reserve of 5%, taking the SCR system into consideration. The engine was, as is likely to be permitted by law, at operating temperature (T.sub.off>70° C.) at the beginning of the RDE cycle.

(34) For case 2, the variant with a minimal NO.sub.x emission was set due to the existing control deviation after passing the minimum distance s.sub.min (for this, see also FIG. 5). As of approximately second 2,600, the NO.sub.x limiting value fell short so that variants can be configured that are favorable with regard to consumption. The operating with high EGR rates at the beginning of the cycle could be influenced in a positive manner due to an adaptation of the control parameters or the expansion of the minimum distance s.sub.min for the initialization range with regard to drivability. As a consequence, a decrease in soot emission would also result.

(35) The exhaust gas aftertreatment at operating temperature for case 3 already made a conversion of nitrogen oxide possible at the beginning of the cycle. Thereby, the most economic variant can be configured up to the phases with a strong vehicle acceleration or at a lower exhaust gas temperature. During the second part of the city phase (1,200 s to 2,400 s), the exhaust gas temperature decreased to under 200° C. due to the distance profile, thereby reducing the implementation level of the SCR system. Consequently, with the aid of higher EGR rates, the raw nitrogen oxide emission was reduced to comply with the NO.sub.x limiting value.

(36) With the aid of the method according to the invention, the advantage resulted that an adaptive adjustment of the respective instantaneous operational state (vehicle, driver, driving profile, etc.) could be achieved in order to comply with the required NO.sub.x limiting value. The potential of the method according to the invention is shown by a variation of vehicle mass and shifting behavior. These tests were carried out for the mentioned RDE cycle with an engine at operating temperature, wherein, for each case, the required limiting value of 80 mg/km including a 5% reserve was complied with.

(37) Preliminary testing for the RDE test cycle has shown that an increase in vehicle mass by approximately 5% causes an increase in CO.sub.2 emission of also approximately 5%. With the aid of the method according to the invention, the higher efficiency of the SCR system can be taken into consideration due to the increase in the exhaust-gas temperature. As a consequence, the increase of fuel consumption can be limited to half as much. However, at the same time, a higher consumption of the dose-injected urea results. A cost-specific optimization with regard to saving fuel and urea consumption is possible.

(38) This applies also to a modified shift strategy. The dynamic shifting behavior with higher gear-shifting speeds reduces the scope of the SCR system. In order to adhere to the NO.sub.x limiting value, the raw nitrogen oxide emission is generally reduced bearing a disadvantage to fuel consumption.

(39) By means of the invention, instantaneously valid and future certification methods for new passenger cars can be taken into consideration. Compliance with the legally prescribed NO.sub.x limiting values during real driving operation is extremely difficult. An adaptation of the engine application up until this point, which covers all possible boundary conditions and factors can only be implemented with great difficulty with regard to the requirements of drivability and low fuel consumption. The proposed transition from the vehicle-individual, almost unchanging engine control system to the adaptive universal regulation, thereby complying with the NO.sub.x limiting value during certification testing, and also in the case of an RDE cycle, is a critical step in the desired direction. According to the vehicle status, driving behavior and environmental conditions, the engine control system can be adapted in a regulated form. At the same time, the method according to the invention, thereby taking the exhaust gas aftertreatment system and the NO.sub.x emission into consideration as a compromise, ensures the best possible driving dynamics and CO.sub.2 emission. As a result, for any engine, only method management is still required in of the instantaneously calculated or measured NO.sub.x value and the vehicle-specific modification that was usual up until this point will be deemed obsolete.