Internal combustion engine and method for controlling the same

Abstract

Internal combustion engine comprising an engine controller (2) for controlling operational parameters of the internal combustion engine (1), wherein the engine controller (2) comprises a model-based controller (3) configured to employ a physical and / or chemical engine model of the internal combustion engine (1) for controlling the operational parameters, wherein a disturbance estimator (4) is provided which is configured to estimate a model error of the engine model based on at least one disturbance quantity, wherein the at least one disturbance quantity comprises at least one of a torque and / or power and / or an angular acceleration and / or a model parameter offset of the internal combustion engine.

Description

[0001] Internal combustion engine and method for controlling the same

[0002] The present invention is concerned with an internal combustion engine according to the classifying portion of claim 1 as well as a method and a computer program product for controlling an internal combustion engine.

[0003] Known in the prior art are highly sophisticated and effective ways of controlling an internal combustion engine. Examples would be EP 2977596 Al, unpublished International Patent Application no. PCT / AT2023 / 060134 and unpublished European Patent Application no. 22209889.

[0004] Such controls for internal combustion engines generally comprise a model-based controller which comprises an engine model for predicting the reaction of the internal combustion engine to the control inputs generated by the engine controller.

[0005] Despite these developments for controlling internal combustion engines less than ideal behaviour of the internal combustion engine can emerge when the engine model on which the predictions or the control is based is not sufficiently accurate for whatever reason.

[0006] In practice, reasons why the employed engine model is not accurate enough may for example stem from an unusual thermal situation (cold start, unusual environment temperature) or from changing fuel composition or quality.

[0007] The object of the invention is to provide an internal combustion engine, a method as well as a method for operating an internal combustion engine, wherein the quality of the control of the internal combustion engines is further increased when the engine model is not accurate, e.g., due to unusual thermal or fuel situations .

[0008] Regarding the internal combustion engine this object is achieved with the features of claim 1, namely by an engine controller for controlling operational parameters of the internal combustion engine, wherein the engine controller comprises a model-based controller configured to employ a physical and / or chemical engine model of the internal combustion engine for controlling the operational parameters, and wherein a disturbance estimator is provided which is configured to estimate a model error of the engine model based on at least one disturbance quantity, and wherein the at least one disturbance quantity comprises at least one of a torque and / or power and / or an angular acceleration and / or a model parameter offset of the internal combustion engine.

[0009] Regarding the method the object of the invention is achieved with the features of claim 12, wherein the method comprises the following:

[0010] - controlling operational parameters of the internal combustion engine using a model-based controller configured to employ a physical and / or chemical engine model of the internal combustion engine,

[0011] - estimating a model error of the engine model based on at least one disturbance quantity, and

[0012] - using at least one of a torque and / or power and / or an angular acceleration and / or a model parameter offset of the internal combustion engine as the at least one disturbance quantity.

[0013] Regarding the computer program product the object of the invention is achieved with the features of claim 13, wherein the computer program product comprises instructions which cause an executing computer to perform the following: - controlling of operational parameters of the internal combustion engine employing a physical and / or chemical engine model of the internal combustion engine,

[0014] - outputting an estimation of a model error of the engine model based on at least one disturbance quantity, and

[0015] - using at least one of a torque and / or power and / or an angular acceleration and / or a model parameter offset of the internal combustion engine as the at least one disturbance quantity.

[0016] The invention makes it possible to control the internal combustion engine such that one or more tracking variables of an internal combustion engine, for example an engine speed, an engine power, an engine torque, an engine boost pressure, engine NOx emissions, an air / fuel equivalence ratio or an egr mass flow does not suffer an offset from a desired value or range of values.

[0017] If the internal combustion engine drives a generator for generating electrical energy, the generator power and / or the generator frequency can be used alternatively or additionally as tracking variable.

[0018] The mechanism to achieve the removal of any offset between the desired value and the measurement of the tracking variable is preferably the correction of the setpoint of these operational parameters to the model-based controller which is computed based on the estimated model error. Compensating the model error by the correction of the setpoint is equivalent to a zero offset in the tracking variable.

[0019] This "zero-offset-tracking"-behaviour is obviously great advantage because it makes sure that the steady state operating point of the internal combustion engine is exactly at the desired point, while it is at the same time possible to use a model-based controller which is known to result in very good short term control behaviour (no or small amounts of oscillations before returning to a steady state after a rapid change in load for example). More generally speaking, the invention makes it possible to control an internal combustion engine extremely well not only in the short term, but also in the long term.

[0020] In particularly preferred embodiments the model-based controller can be configured to minimise a cost function based on predictions of the engine model, wherein the output of the model-based controller, e.g., command values for a boost pressure and / or the air-fuel equivalence ratio, can be determined from the minimisation of the cost function.

[0021] The internal combustion engine can preferably be a gas engine configured to combust molecular hydrogen, natural gas, and / or other hydrocarbons for creating mechanical power.

[0022] The internal combustion engine can for example be piston-cylinder- type internal combustion engine, with any number, preferably an even number, of cylinders.

[0023] Such gas engines can be configured, controlled, or operated as otherwise described in the introductory part of the invention.

[0024] They can use stoichiometric lambda or a lean burn concept.

[0025] They can preferably comprise an exhaust gas recirculation (egr) system.

[0026] The internal combustion engine according to the invention can comprise at least one piston-cylinder unit, a charging system (e.g., a turbo charger, an electrically driven compressor, or mixed forms) with for example one, two, three, or four stages, and / or an exhaust gas aftertreatment system.

[0027] Protections is also sought for a so-called genset which comprises the internal combustion engine according to the invention as well as a generator, which is coupled mechanically to the internal combustion engine for driving the generator and for producing electrical energy.

[0028] Protection is furthermore sought for a transitory or non- transitory data storage device having a computer program product according to claim 11 stored thereon.

[0029] It should be noted that the engine controller can be embodied as one or more software and / or hardware modules.

[0030] The engine controller can be embodied in a distributed fashion where some sub-modules are present or run at the internal combustion engine and others are present or run at a computer distanced from the internal combustion engine, e.g., in a cloud server or the like.

[0031] The computer program product according to the invention can be executed on the engine controller. In other embodiments, the computer program product can for example be used to control a virtual internal combustion engine, for example as part of a simulation of the internal combustion engine.

[0032] Further advantageous embodiments are defined in the dependent claims.

[0033] The engine controller may be configured to shift or offset setpoints of the operational parameters based on the estimated model error for compensating the model error. Formulated differently, the disturbance estimator can be configured to output shifted setpoints for the operational parameters.

[0034] Preferably, the shifted setpoints can be used by the engine controller for control of the engine.

[0035] In preferred embodiments the operational parameters comprise an engine speed, an engine power, a boost pressure, an air-fuel equivalence ratio, emissions (e.g., NOx emission and / or hydrocarbon emission), and / or an exhaust gas recirculation rate.

[0036] In the context of the present invention the air-fuel equivalence ratio can be understood as the ratio of the amount of air to fuel normalised so that the air-fuel equivalence ratio is one when the air fuel ratio is stoichiometric.

[0037] In particularly preferred embodiments the tracking variable can be measured or be determined otherwise and given to the disturbance estimator as feedback variable or variables. This makes it in particular possible to achieve zero-offset-tracking in a relatively simple way.

[0038] The operational parameters can preferably comprise a first operational parameter and a second operational parameter, the engine controller has stored at least one boundary value for the first operational parameter, and the engine controller is configured to shift or offset the second operational parameter if the compensation of the model error would imply offsetting or shifting the first operational parameter beyond the at least one boundary value, preferably while the first operational parameter is saturated to the at least one boundary value. It should be noted that there can be more than one first operational parameter and / or more than one second operational parameter.

[0039] For example, if there are two or more first operational parameters, the second operational parameter can be shifted if all and / or some of the first operational parameters would have to be offset or shifted beyond the at least one boundary value.

[0040] If there are more than one second operational parameters preferably all are shifted or offset if all and / or some of the first operational parameters would have to be offset or shifted beyond the at least one boundary value.

[0041] The first operational parameter can preferably comprise the air- fuel equivalence ratio, and / or the second operation parameter can preferably comprise the boost pressure.

[0042] The first operational parameter can preferably comprise an exhaust gas recirculation rate and / or an exhaust gas recirculation mass flow, and / or the second operational parameter can preferably comprise a boost pressure.

[0043] The engine model according to the invention may be a linear version or a non-linear version of a model of the internal combustion engine.

[0044] In particularly preferred embodiments the disturbance estimator can comprise an augmented engine model which comprises the at least one disturbance quantity, preferably wherein the augmented engine model comprises a linear version of the engine model and / or a non- linear version of the engine model. The model-based controller can preferably comprise lower-level controllers configured to perform lower-level control of actuators, preferably a throttle valve, a compressor bypass valve, a wastegate valve, an ignition system, a fuel metering valve, and / or an exhaust gas recirculation valve and / or a blow-off valve. This can be a preferred way for controlling the boost pressure by actuating the compressor bypass valve, the throttle valve, the wastegate valve, and / or the blow off valve.

[0045] The model-based controller may be configured to output command values, preferably for the boost pressure and / or the air-fuel equivalence ratio, and wherein the lower-level controllers may be configured to use the command values outputted by the model-based controller as reference values for the lower-level control of the actuators .

[0046] Formulated differently, in preferred embodiments there is a cascaded control scheme, and the invention makes sure that the actuators are actuated such that the described zero-offset- tracking from a desired operating point which could be quantified using a tracking variable, which is however not necessary in the context of the invention.

[0047] According to the invention a control can be embodied in different ways, for example as open loop control or closed loop control, with one input and / or one output variable. However, in particular the model-based controller can be a multi-input / multi-output controller .

[0048] The low-level controls can be preferably implemented as open loop control or PID control (potentially with any of the PID gains zero). It furthermore can include models for feedforward or for open loop control that are possibly inverted. Preferred measurement values which can be taken at the internal combustion engine and / or the generator and which can be used in any of the controls of the cascaded control according to the invention as feedback are for example:

[0049] - engine speed and / or equivalent grid frequency and / or

[0050] - generator power and / or crankshaft torque and / or

[0051] - generator voltage and / or

[0052] - boost pressure before and / or after the throttle valve and / or

[0053] - boost temperature (i.e., the temperature of the air and / or air / fuel mixture downstream of a compressor).

[0054] - NOx concentration

[0055] - air-fuel equivalence ratio

[0056] A cascaded control is understood as a control with at least one high-level control and at least one low-level control wherein at least one output variable of the at least one high-level control is used as at least one setpoint for the at least one low-level control .

[0057] For the purposes of this invention the air-fuel equivalence ratio (also called lambda) is the ratio of the masses of air and fuel in the mixture divided by the stoichiometric air / fuel ratio.

[0058] The at least one control unit can comprise a model-based controller, preferably model predictive controller and / or a state space controller, for performing the at least one high-level control, preferably during the load transient operation and / or substantially during all operation.

[0059] Instead of a model predictive controller and / or a state space controller the model-based controller can be another controller configured to output the at least one setpoint for the at least one low-level controller by solving a mathematical optimization problem based on actual measurements and a dynamic engine model to minimize the control error.

[0060] The model-based controller can comprise a cost function to be minimised, wherein the cost function is based on the engine model of the internal combustion engine, potentially together with a generator coupled to the internal combustion engine and / or a power supply grid and / or participants of the power supply grids, such as other generators for creating electrical energy or loads.

[0061] In particular, the model-based controller can be configured for solving an optimization problem that makes use of a dynamic engine model to predict the evolution of relevant engine variables (e.g., speed, boost pressure, power, torque, lambda, EGR concentration) over a finite or infinite prediction horizon in response to the selected setpoints to the low level controllers and therefore is able to coordinate the setpoints in a preferably optimal way (e.g. linear quadratic regulator, model predictive controller).

[0062] The model-based controller can apply nonlinear control techniques such as nonlinear model predictive control, feedback linearization or backstepping, or linear control techniques that work with a model that is linearized around the actual / reference operation point or around the actual / reference trajectory of state / system variables.

[0063] The controller may receive measurements of power that provide the feedforward of the measured disturbance that is responsible for a fast reaction and saturation of the references to the at least one low-level control in case of a load transient. A feedback control based on the speed and boost pressure and / or other measurements can provide the additional control action for stabilizing the system at the desired references. The basis for the model-based controller is preferably a dynamic engine model that describes the evolution of relevant engine variables (speed, boost pressure, air fuel ratio, torque and / or power and possibly others) depending on the control inputs and external disturbances such as electrical loads on the generator that is coupled to the engine.

[0064] An example for such a model is provided by J. Huber, H. Kopecek, and M. Hofbaur. In "Nonlinear model predictive control of an internal combustion engine exposed to measured disturbances", Control Engineering Practice 44, (2015) for the lean burn case without exhaust gas recirculation or as a minimum formulation given by the following equations including EGR.

[0065] A torque balance between engine torque ieand generator load iggoverns the engine speed is given by Guzzella, Lino, and Christopher Onder in "Introduction to modelling and control of internal combustion engine systems", Springer Science & Business Media, (2009) as with the engine torque Tecomputed as depending on volumetric efficiency r)vol, brake efficiency T)brake, displacement volume Vd, lower heating value Hi, gas constant of the mixtureRim, intake manifold temperature Tim, stoichiometric air fuel ratio Lmin, as well as the controlled variables engine speed omega, intake manifold pressure p2', air fuel equivalence ratio A and the exhaust gas recirculation fraction cegr, leading to with

[0066] The following assumptions for the closed loop control of the lower- level controls may be taken for the design of the model-based controller control.

[0067] The tracking behaviour of boost pressure, with boost pressure reference and measurement p2' can be modelled with time constant T and slope K with (see Fig. 6)

[0068] Or further simplified, e.g.,

[0069] The tracking behaviour of air fuel equivalence ratio A can be modelled with a dependence on the delay between the gas dosage / fuel metering and the cylinders D as well as a mixing time constant with

[0070] The tracking behavior of exhaust gas recirculation concentrationcegr can be modelled with a dependence on the transport delay between the EGR valve and the cylinders Degras well as a mixing time constant Tegr,

[0071] Using

[0072] (control variables), (references), values) (commandthe model described above can be denoted in a simpler manner as

[0073] (see also Fig. 1)

[0074] There may be many sources of uncertainties for these models. The applied gas type can change, with uncertainty in its actual composition and the related lower heating value Hi and stoichiometric air fuel equivalence ratio Lmin. On the other hand, the engine setup itself may be quite uncertain, with a large number of possible manufacturing variants characterized for example by different compression ratios, turbocharger layouts and piston types influencing the volumetric and brake efficiencies J]vo(and r / brake• Furthermore, application specific parameterization of ignition timing, cooler temperatures etc. influence the same quantities. Of course, also aging and service actions that influence the static and dynamic behavior of the engine can introduce another source of uncertainty. According to the invention these and other types of uncertainties can be addressed.

[0075] According to the invention a disturbance estimator is used for estimating at least one disturbance quantity. In one example embodiment the disturbance estimator comprises an augmented engine model which comprises the at least one disturbance quantity. An example of such an augmented engine model is given by the following equation (augmented engine model equation):

[0076] Here,

[0077] As is evident the augmented engine model equation is an extension of the engine model equation containing a disturbance scalar quantity d. In other embodiments the disturbance quantity could be multi-dimensional . d in this example is an angular acceleration, i.e., a torque scaled by the moment of inertia of the rotating parts of the genset.

[0078] The design of the observer gain matrix [Lx,Ld] is done by a Kalman Filter approach, using weighting matrices QKFand RKF. Even if the engine model uses a nonlinear model, the recursive Kalman Filter design can be performed using the linearized version of the augmented engine model equation.

[0079] The feedback variables from the engine are available measurements of states variables x(fc) such as engine speed, boost pressure and lambda. If certain measurements are not available these quantities can be estimated in addition to the disturbance quantities (e.g., if no lambda sensor is available).

[0080] In this embodiment it is envisioned that the setpoint for engine speed and boost pressure are not shifted, and that - initially - only the setpoint for the air-fuel equivalence ratio is shifted. Therefore, and

[0081] For an example of the LEANOX module the reader is referred to the description of Fig. 2.

[0082] The condition that the engine speed, i.e., Xi, should not change can be expressed mathematically by the steady state solution of the differential equation of the engine speed, e.g., setting the derivative equal to zero (i)e= x1= 0, which leads to

[0083] From this condition the shifted lambda setpoint can be computed as

[0084] In the present embodiment there is a lower limit x3fminand an upper limit x3,maxfor the air-fuel equivalence ratio. If the setpoint for the air-fuel equivalence ratio reaches one of these limits the setpoint for the air-fuel equivalence ratio is not shifted beyond the mentioned limits.

[0085] Rather, in such situations the setpoint for the boost pressure is shifted. The new boost pressure setpoint can also be determined from the condition given above to yield

[0086] In this embodiment the air-fuel equivalence ratio is therefore the first operational parameter and the boost pressure is the second operational parameter as mentioned above.

[0087] In other embodiments it is of course possible to for example use the boost pressure as the first operational parameter and the air- fuel equivalence ratio as the second operational parameter, or more complex engine models or augmented engine models could be used.

[0088] Further details and advantages of the invention are apparent from the figures and the accompanying description of the figures. The figures show:

[0089] Fig. 1 a schematic depiction of an embodiment of a genset according to the invention, Fig. 2 simulation results showing the effect of the invention,

[0090] Fig. 3 a schematic depiction of an embodiment of an internal combustion engine according to the invention, and

[0091] Fig. 4 a schematic depiction of an embodiment of a genset according to the invention.

[0092] Fig. 1 shows a schematic depiction of an embodiment of a genset 10 according to the invention.

[0093] The engine controller 2 comprises a model-based controller 3 and a disturbance estimator 4. Functions of the model-based controller 3 and the disturbance estimator 4 have already been described above in connection with the engine model equation, the augmented engine model equation and the remarks following thereafter.

[0094] The possibly shifted setpoints for the engine speed, the boost pressure, and the air-fuel equivalence ratio are outputted by the disturbance estimator 4 and used as input for the model-based controller 3.

[0095] Additionally, the estimated (or directly measured) values for these quantities, denoted by a hat, are also given to the model- based controller 2 and are used in the model-based controller 3 as a state feedback to control the engine to the references.

[0096] Additionally, the engine controller 2 comprises in this embodiment a LEANOX module which sets a reference boost pressure P2,ref' based on the estimated power or load of the genset so as to achieve a desired level of NOx emissions.

[0097] An embodiment for how the LEANOX module functions can be found in EP 2977596 Al. Additionally, the engine controller 2 comprises in this embodiment a torque balance module. This module solves the model equations for the engine model for a steady state condition at the actual load estimate, by setting the derivatives equal to zero (e.g.,

[0098] The model-based controller 2 outputs command values for the boost pressure (uP2') and for the air-fuel equivalence ratio (ux).

[0099] The model-based controller 3 comprises lower-level controllers 5 configured to perform lower-level control of actuators, preferably a throttle valve 6, a compressor bypass valve 7, a wastegate valve 8, an ignition system 9, a fuel metering valve 11, and / or an exhaust gas recirculation valve 12.

[0100] The valves can for example be controlled in their opening position or opening degree. Alternatively, the valves can be only controlled according to a desired (completely) closed or (completely) open position .

[0101] The ignition system 9 can be controlled regarding the ignition timing, i.e., determining when or at which crank angle the ignition of the combustion takes place.

[0102] The model-based controller 3 is configured to output the command values for the boost pressure and the air-fuel equivalence ratio.

[0103] The lower-level controllers 5 are configured to use these command values as reference values for the lower-level control of the actuators . Therefore, in this embodiment there is a cascaded control scheme, and the invention makes sure that the actuators are actuated such that the described zero-offset-tracking from a desired operating point is achieved, in this case characterised by a desired window for the engine speed (see Fig. 3b and 4) as a tracking variable.

[0104] It should be noted that in the depiction of Fig. 1 the lower-level controllers 5 are depicted as one box together with the static mechanical parts of the internal combustion engine 1. As is completely clear for persons skilled in the art the lower-level controllers 5 can be embodied as hardware and / or software modules which are run and / or arranged together or separately from each other, e.g., separate software modules executed on the same or different computers or separate hardware modules arranged at different locations at or near the internal combustion engine 1.

[0105] The internal combustion engine 1 drives a crankshaft and other rotating parts which are in turn coupled mechanically to a rotor of a generator 13 which is configured to generate electrical energy which can for example be delivered to an energy distribution grid or, for short, power grid 19.

[0106] An engine speed sensor 20 (see Fig. 5) is used to determine the speed of the crankshaft and / or the rotor of the generator 13 which is coupled mechanically with the crankshaft of the internal combustion engine 1. The measured engine speed is given to the disturbance estimator 4 as a feedback variable.

[0107] It should be noted that the generator power PGand generator voltage V, which is used for estimating the engine load or power by the load estimator module, is not necessarily an input variable for the model-based controller 3 or the disturbance estimator 4. As such, it is clear that the invention can also be implemented in internal combustion engines 1 which are not coupled mechanically to a generator 13.

[0108] Fig. 2 shows results from a simulation of an internal combustion engine 1 together with a generator 13 operated according to the invention. As can be seen from the diagram on the upper left the engine speed tracking is excellent even though there are numerous load steps as is evident from the diagram in the upper right.

[0109] Note that at time ~100sec until time ~170sec, the lambda command (lower right diagram) is always saturated, and the steady state speed tracking is instead fulfilled by shifting the setpoint for the boost pressure. Briefly, this also happens at the 250s and 300s marks.

[0110] It is also noted that the computed lambda (Xest) vales follow the command value for lambda (Xcmd) so closely that both graphs are indistinguishable .

[0111] Fig. 3 shows a schematic depiction of an embodiment of an internal combustion engine 1 according to the invention.

[0112] A fuel metering valve 11 is provided for mixing a fuel and air to produce a combustible air / fuel mixture.

[0113] The fuel can for example be natural gas including methane and / or molecular hydrogen. Alternatively, or additionally, the fuel may comprise other hydrocarbons.

[0114] The air / fuel mixture is compressed in compressor 16 of the turbocharger 15 so that the air / fuel mixture is charged into a piston-cylinder unit 14 or a plurality of piston-cylinder units 14 while under a boost pressure p2f(potentially together with recirculated exhaust gas, see below).

[0115] In a bypass conduct bypassing the compressor 16 of the turbocharger 15 there is a compressor bypass valve 7 which can be used to direct an amount of the air / fuel mixture around the compressor 16 so as to lower the boost pressure p2f.

[0116] In the conduct connecting the compressor 16 of the turbocharger 15 to the piston-cylinder unit(s) 14 there is a throttle valve 6.

[0117] In this embodiment the engine is mixture charged as the fuel metering valve 11 is upstream of the compressor 16. In other embodiments the fuel metering valve 11 can be arranged downstream of the compressor 16 (air charged engine).

[0118] The turbocharger 15 in this example is a single stage turbocharger 15. In other embodiments according to the invention there can two, three, four or more turbocharger 15 stages.

[0119] The internal combustion engine 1 could also include a blowoff valve for rapidly discharging charged air and / or charged air / fuel mixture into the environment or other separate volumes. However, in this embodiment such a blowoff valve is not included.

[0120] The cylinder charge, which is the air / fuel mixture charged into the piston-cylinder unit(s) 14 under the boost pressure p2fis ignited using an ignition system 9, in this case a system comprising a spark plug for each piston-cylinder unit 14.

[0121] In principle, the invention can also be used with compression ignition engines and / or engines operated with liquid fuel and / or dual fuel engines. The piston-cylinder unit 14 can comprise a pre-combustion chamber.

[0122] After combustion in the piston-cylinder unit(s) 14 the exhaust gases remaining in the cylinder are expelled therefrom.

[0123] There is an exhaust gas recirculation passage in which an exhaust gas recirculation valve 12 is arranged.

[0124] The exhaust gas recirculation valve 12 can be used to recirculate part of the exhaust gas into the mass flow of the air / fuel mixture directed into the piston-cylinder unit 14, such that the cylinder charge comprises recirculated exhaust gas next to the air / fuel mixture .

[0125] The exhaust gas which is not recirculated is decompressed in the exhaust gas turbine 17 of the turbocharger 15. The exhaust gas turbine 17 drives the turbocharger shaft 18 which in turn drives the compressor 16 of the turbocharger 15.

[0126] Alternatively, or additionally, to the exhaust gas turbine 17 the charging system can comprise an electric drive for the compressor 16 of the turbocharger 15.

[0127] There is a bypass conduct bypassing the exhaust gas turbine 20 and a waste gate valve 8 is arranged in this bypass of the exhaust gas turbine 17 so that part of the exhaust gas can be routed past the exhaust gas turbine 17.

[0128] The exhaust gas passing through the exhaust gas turbine 17 and / or the wastegate valve 8 is then subjected to aftertreatment in the exhaust gas aftertreatment system 21 which can comprise different kinds of catalytic converters. For example, if the internal combustion engine 1 is operated with essentially stoichiometric lambda the exhaust gas aftertreatment system 21 can comprise a three-way catalytic converter.

[0129] In other embodiments where the internal combustion engine 1 is operated with a lean burn concept the exhaust gas aftertreatment system 21 can comprise a selective catalytic reaction catalytic converter and / or an oxidation catalytic converter and / or a thermal oxidiser.

[0130] In embodiments where the internal combustion engine 1 is operated with a lean burn concept the exhaust gas recirculation valve 12 and the corresponding conduct may not be present.

[0131] In embodiments where the internal combustion engine can be operated both stoichiometrically and with a lean burn concept as desired the exhaust gas recirculation valve 12 can preferably be kept closed as long as the internal combustion engine 1 is operated at lambda greater than one (lean operation).

[0132] In other embodiments the exhaust gas recirculation valve 12 or the exhaust gas recirculation system is not present.

[0133] The engine controller 2 is in signal communication with the actuators of the internal combustion engines, which in this embodiment comprise the fuel metering valve 11, the exhaust gas recirculation valve 12, the ignition system 9, the compressor bypass valve 7, the throttle valve 6, and the wastegate valve 8.

[0134] However, the signal communication of the control unit 2 with the actuators is not depicted in Fig. 3 for the sake of clarity of the figure. The engine controller 2 of this embodiment is also in signal communication with a multitude of measurement devices which can be used to implement closed loop control, both for the model-based controller 3 and the lower-level controllers 5.

[0135] Measurement devices which can for example be used in this capacity are

[0136] - pressure sensors, upstream and / or downstream of the compressor 16 and / or upstream and / or downstream of the exhaust turbine 17, and / or

[0137] - temperature sensors, upstream and / or downstream of the compressor 16 and / or upstream and / or downstream of the exhaust turbine 17, and / or

[0138] - in cylinder pressure sensors and / or

[0139] - knock sensors and / or

[0140] - lambda sensors in the exhaust gas conduct and / or

[0141] - oxygen concentration sensors and / or

[0142] - engine speed sensors

[0143] - NOx sensors

[0144] These sensors can be embodied as in principle known in the prior art.

[0145] Also here, the signal communication of the engine controller 2 with the sensors is not depicted in Fig. 3 for the sake of clarity the figure.

[0146] In the embodiment of Fig. 3 the exhaust gas recirculation valve 12 recirculates exhaust gas from downstream of the exhaust gas turbine 17 to a volume upstream of the compressor 16 when at least partially opened. This is called low-pressure exhaust gas recirculation. In other embodiments the exhaust gas is recirculated from upstream of the turbine 17 to downstream of the compressor 16, i.e., a low- pressure exhaust gas recirculation. Fig. 4 schematically shows a genset 10 with an internal combustion engine 1 according to the invention which is coupled mechanically to a generator 13 for creating electrical energy.

[0147] The generator 13 may be connected to a power grid 19. The power grid can for example be a public power supply grid, an island grind, or a microgrid.

[0148] List of reference numerals:

[0149] 1 internal combustion engine

[0150] 2 engine controller

[0151] 3 model-based controller

[0152] 4 disturbance estimator

[0153] 5 lower-level controllers

[0154] 6 throttle valve

[0155] 7 compressor bypass valve

[0156] 8 wastegate valve

[0157] 9 ignition system

[0158] 10 genset

[0159] 11 fuel metering valve

[0160] 12 exhaust gas recirculation valve

[0161] 13 generator

[0162] 14 cylinder

[0163] 15 turbo charger

[0164] 16 compressor

[0165] 17 turbine

[0166] 18 turbo shaft

[0167] 19 power grid

[0168] 20 engine speed sensor

[0169] 21 exhaust gas aftertreatment system

Claims

Claims1. Internal combustion engine comprising an engine controller (2) for controlling operational parameters of the internal combustion engine (1), wherein the engine controller (2) comprises a model-based controller (3) configured to employ a physical and / or chemical engine model of the internal combustion engine (1) for controlling the operational parameters, characterised in that a disturbance estimator (4) is provided which is configured to estimate a model error of the engine model based on at least one disturbance quantity, wherein the at least one disturbance quantity comprises at least one of a torque and / or power and / or an angular acceleration and / or a model parameter offset of the internal combustion engine.

2. Internal combustion engine according to claim 1, wherein the engine controller (2) is configured to shift or offset setpoints of the operational parameters based on the estimated model error for compensating the model error.

3. Internal combustion engine according to one of the preceding claims, wherein the operational parameters comprise an engine speed, an engine power, a boost pressure, an air-fuel equivalence ratio, emissions, and / or an exhaust gas recirculation rate.

4. Internal combustion engine according to one of the preceding claims, wherein the engine controller (2) is configured to receive a measurement of a tracking variable and / or to determine a value of a tracking variable, and wherein the disturbance estimator (4) is configured to use the measured ordetermined tracking variable as feedback for updating the estimated model error.

5. Internal combustion engine according to one of the preceding claims, wherein the operational parameters comprise a first operational parameter and a second operational parameter, the engine controller has stored at least one boundary value for the first operational parameter, and the engine controller is configured to shift or offset the second operational parameter if the compensation of the model error would imply offsetting or shifting the first operational parameter beyond the at least one boundary value, preferably while the first operational parameter is saturated to the at least one boundary value.

6. Internal combustion engine according to claim 5, wherein the first operational parameter comprises the air-fuel equivalence ratio, and / or the second operational parameter comprises the boost pressure.

7. Internal combustion engine according to claim 5 or 6, wherein the first operational parameter comprises an exhaust gas recirculation rate and / or an exhaust gas recirculation mass flow, and / or the second operational parameter comprises a boost pressure.

8. Internal combustion engine according to one of the preceding claims, wherein the disturbance estimator (4) comprises an augmented engine model which comprises the at least one disturbance quantity, preferably wherein the augmented engine model comprises a linear version of the engine model and / or a non-linear version of the engine model.

9. Internal combustion engine according to one of the preceding claims, wherein the model-based controller (3) compriseslower-level controllers (5) configured to perform lower-level control of actuators (6,7,8,9,11,12), preferably a throttle valve (6), a compressor bypass valve (7), a wastegate valve (8), an ignition system (9), a fuel metering valve (11), and / or an exhaust gas recirculation valve (12).

10. Internal combustion engine according to claim 9, wherein the model-based controller (3) is configured to output command values, preferably for the boost pressure and / or the air-fuel equivalence ratio, and wherein the lower-level controllers (5) are configured to use the command values outputted by the model-based controller (3) as reference values for the lower- level control of the actuators (6,7,8,9,11,12).

11. Genset comprising an internal combustion engine (1) according to one of the preceding claims and a generator (13) which is coupled mechanically to the internal combustion engine (1) for driving the generator (13) and for producing electrical energy.

12. Method for controlling an internal combustion engine, in particular according to one of the claims 1 to 10, wherein the method comprises the following:- controlling operational parameters of the internal combustion engine (1) using a model-based controller (3) configured to employ a physical and / or chemical engine model of the internal combustion engine (1), and- estimating a model error of the engine model based on at least one disturbance quantity, characterised by using at least one of a torque and / or power and / or an angular acceleration and / or a model parameter offset of the internal combustion engine (1) as the at least one disturbance quantity.

13. Computer program product for controlling, in particular according to a method according to claim 12, an internal combustion engine (1), in particular according to one of the claims 1 to 10, comprising instruction which cause an executing computer to perform the following:- controlling of operational parameters of the internal combustion engine (1) employing a physical and / or chemical engine model of the internal combustion engine (1), and- outputting an estimation of a model error of the engine model based on at least one disturbance quantity, characterised by using at least one of a torque and / or power and / or an angular acceleration and / or a model parameter offset of the internal combustion engine (1) as the at least one disturbance quantity.

14. Transitory or non-transitory data storage device having a computer program product according to claim 13 stored thereon.

Images