Control device and method for operating an internal combustion engine and internal combustion engine with such a control device

Abstract

Control device (3) for operating an internal combustion engine (1), with - a detection module (7) configured to detect a detected emission parameter value (e r ) of the internal combustion engine (1) over a first observation parameter interval from a previous observation parameter value (15) to a current observation parameter value (x a ) to obtain, - a calculation module (9) which is arranged to calculate at least one predicted emission parameter value (e p ) of the internal combustion engine (1) for a second observation parameter interval from the current observation parameter value (x a) up to a prediction observation parameter value (17) such that an emission parameter value of the internal combustion engine (1) averaged over a test interval (27) comprising the first observation parameter interval and the second observation parameter interval has a predetermined emission parameter limit value (e max ) and - a control module (11) which is designed to control the internal combustion engine (1) in accordance with the predicted emission parameter value (e p ) to operate.

Description

[0001] The invention relates to a control device and a method for operating an internal combustion engine and an internal combustion engine with such a control device.

[0002] It is well known that an internal combustion engine, especially an internal combustion engine in a motor vehicle—a passenger car or a truck—must meet legal requirements regarding engine emissions. These emissions are measured in real field operation and must be below a certain limit value on an integral average basis. Depending on the application, this type of emissions calculation is known as real-driving emissions (RDE), in-field confirmity, or in-service monitoring.

[0003] K. Harder et al., "A real-time nonlinear MPC scheme with emission constraints for heavy-duty diesel engines," 2017 American Control Conference (ACC), Seattle, WA, 2017, pp. 240-245, describes a model predictive control (MPC) approach to keep current emissions below the limit. The disadvantage is that this model predictive control is more restrictive than required by law, thus leading to severely limited engine dynamics.

[0004] German patent application DE 10 2016 223 865 A1 discloses a method in which RDE requirements are met by interpolating the setting parameters resulting from different internal combustion engine parameters – particularly with regard to fuel consumption, nitrogen oxide emissions, and engine dynamics. The disadvantage of this is that this adds further complexity to a conventional map-based internal combustion engine control system, particularly in addition to the already very high parameter complexity.

[0005] DE 10 2015 225 279 A1 discloses a method for the predictive control and regulation of an internal combustion engine. DE 10 2018 122 976 A1 discloses a method for evaluating data from at least one measurement of an internal combustion engine in the context of exhaust emission tests under real operating conditions, as well as a control system.

[0006] The invention is therefore based on the object of providing a control device and a method for operating an internal combustion engine and an internal combustion engine with such a control device, wherein the disadvantages mentioned are at least reduced, preferably do not occur.

[0007] The object is achieved by providing the present technical teaching, in particular the teaching of the independent claims as well as the preferred embodiments disclosed in the dependent claims and the description.

[0008] The object is achieved by providing a control device for operating an internal combustion engine. The control device comprises a detection module, a computing module, and a control module. The detection module is configured to obtain a detected emission parameter value of the internal combustion engine over a first observation parameter interval from a previous observation parameter value to a current observation parameter value.The calculation module is configured to determine at least one predicted emission parameter value of the internal combustion engine for a second observation parameter interval from the current observation parameter value to a prediction observation parameter value such that an emission parameter value of the internal combustion engine averaged over a test interval comprising the first observation parameter interval and the second observation parameter interval does not exceed a predetermined emission parameter limit. The control module is configured to operate the internal combustion engine in accordance with the predicted emission parameter value.Advantageously, the control device, in particular the computing module, adjusts the at least one predicted emission parameter value such that an observation parameter average, in particular a time average, for the emission parameter in the test interval does not exceed the emission parameter limit, in particular the statutory limit. Advantageously, the control device thus takes the emission parameter limit into account, in particular as required by law, directly—in particular during operation of the internal combustion engine—and not indirectly.

[0009] The control device is particularly configured to operate the internal combustion engine. In the context of the present technical teaching, operating the internal combustion engine is understood to mean, in particular, controlling or regulating, preferably regulating, the operation of the internal combustion engine.

[0010] In the context of the present technical teaching, a module is generally understood to mean, in particular, a conceptually or physically definable or delimited functional unit configured to perform at least one specific function. This may be a separate computing device, a part of a computing device, a hardware structure, or a software structure, each configured and intended to fulfill the at least one specific function.

[0011] In particular, the current observation parameter value is the observation parameter value for which the most recently recorded emission parameter value is available. Alternatively or additionally, the current observation parameter value is the instantaneous value of the quantized or discrete observation parameter in which the calculations of the calculation module are currently being performed.

[0012] In one embodiment, the detection module is configured to calculate the emission parameter value, in particular based on a simulation and / or a model. For this purpose, the detection module preferably has a virtual sensor or is itself designed as a virtual sensor.

[0013] Alternatively or additionally, the detection module has an interface configured to receive the detected emission parameter value, in particular as a measured value from a physical sensor. The detected emission parameter value is then, in particular, a measured emission parameter value. It is also possible for the detection module to be configured to determine, in particular calculate, the emission parameter value from at least one measured value from a physical sensor.

[0014] In particular, the calculation module is configured to calculate the at least one predicted emission parameter value using the equation 1X(∫x=xa−X+xpxaer(x)dx+xp∑j=1m1mepj)≤emax for m ∈ {1,..., M}, where X is a length of the test interval, x is the observation parameter, x a the current observation parameter value, with x p a length of the second observation parameter interval, with e r the recorded emission parameter value, in particular a recorded emission parameter trajectory, with e p the at least one predicted emission parameter value, with e max is the predetermined emission parameter limit value and M denotes a number of support points within the second observation parameter interval. In particular, for m = 1, one, in particular exactly one, prediction observation parameter value is determined for the second observation parameter interval and equation (1) simplifies to 1X(∫x=xa−X+xpxaer(x)dx+xpep)≤emax.

[0015] Furthermore, for m > 1, a sequence of predicted emission parameter values—consecutive along the observation parameter—is determined for the second observation parameter interval. In particular, it is possible for the predicted emission parameter values ​​to be determined with greater frequency, i.e., in a shorter temporal sequence, than the recorded emission parameter values. In one embodiment, the calculation module is configured to determine the at least one predicted emission parameter value directly as a parameter value.Alternatively, the calculation module is configured to determine the at least one predicted emission parameter value as a function of at least one control value, in particular a plurality of control values, wherein the calculation module is configured in particular to vary the at least one control value, in particular the plurality of control values, such that the predicted emission parameter value calculated on the basis of the function satisfies equation (1) or (2).

[0016] In particular, the control device - in particular the computing module and the control module - is configured to use as a control value at least one parameter selected from a group consisting of an introduction time for introducing, in particular jetting or injecting, a fuel, an introduction quantity of the fuel, an introduction pressure, a fresh air mass flow, valve control times for gas exchange valves, in particular intake and exhaust valves, such as in particular a valve opening time and a valve closing time, an internal combustion engine speed and a combination of at least two of the previous parameters.

[0017] According to a further development of the invention, the calculation module is additionally configured to a) set the previous observation parameter value as an old test interval start value and the prediction observation parameter value as an old test interval end value. Furthermore, the calculation module is configured to b) shift the test interval, starting from the old test interval start value, by a predetermined prediction interval length of the second observation parameter interval toward larger observation parameter values, so that the test interval has a new test interval start value and a new test interval end value.In addition, the calculation module is configured to c) determine a new predicted emission parameter value of the internal combustion engine for a new second observation parameter interval from the old test interval end value to the new test interval end value such that the emission parameter value of the internal combustion engine averaged over the shifted test interval does not exceed the predetermined emission parameter limit. Furthermore, the calculation module is configured to d) repeat steps b) to c) with the new test interval start value as the old test interval start value and the new test interval end value as the old test interval end value until a termination criterion is met.Thus, the control device - in particular the computing module - iteratively shifts the test interval towards larger observation parameter values, so that the predicted emission parameter values ​​are advantageously more finely resolved with respect to the observation parameter value and / or are effectively determined towards larger observation parameter values, i.e. in particular towards the future.

[0018] In particular, the calculation module is configured to shift the test interval, starting from the old test interval start value, into the future by a predetermined prediction interval length of the second observation parameter interval when time is used as the observation parameter. Alternatively, the calculation module is configured to shift the test interval, starting from the old test interval start value, by a predetermined prediction interval length of the second observation parameter interval toward a larger, i.e., more distant, distance—and thus effectively also toward the future—when a distance or a route is used as the observation parameter.

[0019] Particularly preferably, the calculation module is configured to use a maximum number of iterations as the termination criterion. The calculation module is configured to repeat steps b) to c) until the maximum number of iterations is reached.

[0020] In one embodiment, the maximum number of iterations is based on a predetermined temporal prediction horizon and the predetermined prediction interval length, wherein a time interval between the current observation parameter value and the new test interval end value is less than or equal to the predetermined temporal prediction horizon. Particularly preferably, the maximum number of iterations is the largest integer that is smaller than the quotient of the predetermined temporal prediction horizon and the predetermined prediction interval length.

[0021] Particularly preferably, the control device—in particular the control module—is configured to operate the internal combustion engine in accordance with the predicted emission parameter value after the termination criterion has been reached. Alternatively or additionally, the control device—in particular the control module—is configured to operate the internal combustion engine in accordance with the predicted emission parameter value of the old test interval, while the calculation module calculates the predicted emission parameter value of the new test interval.

[0022] In particular, the control device - in particular the computing module - is configured to reach the termination criterion after a maximum of 10 ms, particularly preferably after a maximum of 5 ms.

[0023] In particular, the calculation module is set up to iteratively calculate at least N predicted emission parameter values ​​using the equation 1X(∫x=xa−X+xpxaer(x)dx+xp∑i=1n1mepj,j)≤emax for m ∈ {1,..., M} and n = 1,..., N, where N denotes a number of iterations - in particular a number of second observation parameter intervals. In particular, for m = 1, one, in particular exactly one, prediction observation parameter value is determined for each of the second observation parameter intervals - thus a total of N predicted emission parameter values ​​- and equation (3) simplifies to 1X(∫x=xa−X+n⋅xpxaer(x)dx+xp∑i=1nepj)≤emax.

[0024] Furthermore, for m > 1, a sequence of predicted emission parameter values ​​- consecutive along the observation parameter - a total of N · M predicted emission parameter values ​​- is determined for each of the second observation parameter intervals. In one embodiment, the calculation module is configured to determine the predicted emission parameter values ​​directly as parameter values. Alternatively, the calculation module is configured to determine the predicted emission parameter values ​​as functions of the at least one control value, in particular of the plurality of control values, wherein the calculation module is configured in particular to vary the at least one control value, in particular the plurality of control values, such that the predicted emission parameter values ​​calculated on the basis of the functions each satisfy equations (3) or (4).

[0025] According to a further development of the invention, the calculation module is configured to calculate the at least one predicted emission parameter value using a Gaussian process model. Gaussian process models are particularly suitable for modeling and / or controlling emissions from an internal combustion engine: Compared to polynomial-based models, they are particularly easy to adapt to new or changed data points in the application field, and they exhibit more suitable and physically more correct behavior in the boundary regions of the given parameter space. Compared to physical models, they require significantly less computational effort. Furthermore, they enable the direct use of test bench data. Such a Gaussian process model is particularly given by stored data points (X) obtained, for example, in test bench experiments. b , Y b ), where X b ∈ ℝ n×min particular n input variables for m different operating states and with Y b ∈ ℝ m×k In particular, k output variables are specified for the m different operating states. In particular, the input variables X b a subset of the union of the at least one control value, in particular the plurality of control values, the internal combustion engine, and at least one internal combustion engine model parameter. The output variables Y b are a subset of at least one emission parameter. Furthermore, the Gaussian process model is characterized by a predetermined calculation scheme for an expected value E(X u ) ∈ ℝ l×k and a variance Var(X u ) for input variables not included in the original data set for / different operating states X u ∈ ℝ n×l given: E(Xu)=m(Xu)+K(Xu,Xb)(K(Xb,Xb)+σ2I)−1(Yb−m(Xb)), Var(Xu)=K(Xu,Xu)+σ2−K(Xu,Xb)(K(Xb,Xb)+σ2I)−1K(Xb,Xu), with a mean function m(X u ), a predetermined variance σ 2 , the identity matrix I, and a covariance function K, which depends on the Euclidean distance r between two points x1, x2 in the following way: K(X1,X2)=(k(X1:n,i1,X1:n,j2))i=1,…,m,j=1,…,m, k(x1,x2)=σF2exp(−r(x1,x2)22l2), with a predetermined distance parameter l and a predetermined signal variance σ F . Thus, in equations (5) and (6) K(X u , X b ) ∈ ℝ l×m K(X b , X b ) ∈ ℝ m×m I ∈ ℝ m×m and Y b ∈ ℝ m×k .

[0026] The mean function m(x) is preferably obtained as a Gaussian process model.

[0027] In particular, a first Gaussian process model, also referred to as the basic grid, is first fitted to second test bench data under at least one constraint derived from the first test bench data. In particular, input variables X b selected, and the corresponding output variables Y b are calculated in such a way that a deviation of the expected value E(X) of the first Gaussian process model, which is determined by the input variables X b and the output variables Y bis determined, is minimized to the second test bench data while observing the constraint. Furthermore, for the purpose of determining the first Gaussian process model, it is preferably assumed that its mean value function m(x) = 0. The first test bench data comprise a larger parameter space than the second test bench data. In particular, it is possible for the first test bench data to be measured on a single-cylinder test bench, while the second test bench data are measured on the full engine or likewise on the single-cylinder test bench and, in the latter case, are preferably converted to the full engine using a simulation model. The constraint is preferably obtained as a trend, wherein, for example, it is determined whether certain parameters behave linearly or monotonously with respect to one another.If no such trend is observed, the constraint can be omitted, whereby the fit of the Gaussian process model to the second test bench data is then also referred to as unconstrained.

[0028] The expected value of the first Gaussian process model thus obtained is then used in a next step as the mean function m(x) in a second Gaussian process model, into which the second test bench data are now input as known input variables X b2 and output variables Y b2 enter into.

[0029] According to a further development of the invention, the control device has an operating plan module configured to determine, based on the at least one predicted emission parameter value, the at least one control value, in particular a plurality of control values, for operating the internal combustion engine. Alternatively or additionally, the operating plan module is configured to use the Gaussian process model to determine the at least one control value, in particular the plurality of control values, for operating the internal combustion engine. Advantageously, the control device thus determines the at least one control value, in particular the plurality of control values, for particularly optimal operation of the internal combustion engine.

[0030] Preferably, the operating plan module is configured to create an operating plan of the internal combustion engine based on the at least one control value, in particular the plurality of control values.

[0031] In a particularly preferred embodiment, the operating plan module is configured to determine the at least one control value, in particular the plurality of control values, for operating the internal combustion engine based on a model predictive control algorithm (MPC algorithm) and / or an internal combustion engine model. In particular, the operating plan module is configured to solve a general dynamic optimization problem of the MPC algorithm, taking into account one of equations (1) to (4) as a constraint, and thus to determine the at least one control value, in particular the plurality of control values. Advantageously, the control device is configured to operate the internal combustion engine in such a way that the emissions limit, in particular the statutory limit, is complied with while simultaneously maintaining good internal combustion engine dynamics.

[0032] According to a further development of the invention, the control module is configured to operate the internal combustion engine based on the at least one control value, in particular the plurality of control values.

[0033] In particular, the control device is configured to implement the method according to the invention explained below or a method according to one or more of the embodiments explained below. In particular, the description of the control device, on the one hand, and the method, on the other, are to be understood as complementary.

[0034] The object is also achieved by providing an internal combustion engine with a control device according to the invention or a control device according to one or more of the previously explained embodiments. In connection with the internal combustion engine, the advantages already explained in connection with the control device are particularly evident.

[0035] According to a further development of the invention, the internal combustion engine has an emissions sensor which is configured to determine the at least one emissions parameter value.

[0036] The object is also achieved by providing a method for operating an internal combustion engine. A detected emission parameter value of the internal combustion engine is detected over a first observation parameter interval from a past observation parameter value to a current observation parameter value. Furthermore, at least one predicted emission parameter value of the internal combustion engine is determined for a second observation parameter interval from the current observation parameter value to a prediction observation parameter value such that an emission parameter value of the internal combustion engine averaged over a test interval comprising the first observation parameter interval and the second observation parameter interval does not exceed a predetermined emission parameter limit value. The internal combustion engine is operated in accordance with the predicted emission parameter value.In connection with the method, the advantages arise in particular which have already been explained in connection with the control device and the internal combustion engine.

[0037] In particular, the current observation parameter value is the observation parameter value for which the most recently recorded emission parameter value is available. Alternatively or additionally, the current observation parameter value is the instantaneous value of the quantized or discrete observation parameter in which the calculations are currently being performed.

[0038] In one embodiment, the emission parameter value is calculated, in particular, based on a simulation and / or a model. Alternatively, the recorded emission parameter value is received.

[0039] In particular, the at least one predicted emission parameter value is calculated using equation (1) or (2). In one embodiment, the at least one predicted emission parameter value is determined directly as a parameter value. Alternatively, the at least one predicted emission parameter value is determined as a function of at least one control value, in particular a plurality of control values, wherein the at least one control value, in particular the plurality of control values, is varied such that the predicted emission parameter value calculated using the function satisfies equation (1) or (2).

[0040] In particular, at least one parameter is used as the control value, which is selected from a group consisting of an introduction time for introducing, in particular jetting or injecting, a fuel, an introduction quantity of the fuel, an introduction pressure, a fresh air mass flow, valve control times for gas exchange valves, in particular intake and exhaust valves, such as in particular a valve opening time and a valve closing time, an internal combustion engine speed and a combination of at least two of the previous parameters.

[0041] According to a further development of the invention, it is provided that a) the previous observation parameter value is used as an old test interval start value and the prediction observation parameter value is used as an old test interval end value. Furthermore, b) the test interval is shifted from the old test interval start value by a predetermined prediction interval length of the second observation parameter interval in the direction of larger observation parameter values, so that the test interval has a new test interval start value and a new test interval end value. Subsequently, c) a new predicted emission parameter value of the internal combustion engine is determined for a new second observation parameter interval from the old test interval end value to the new test interval end value such that the emission parameter value of the internal combustion engine averaged over the shifted test interval does not exceed the predetermined emission parameter limit value.In addition, d) steps b) to c) are repeated with the new test interval start value as the old test interval start value and the new test interval end value as the old test interval end value until a termination criterion is reached. This iteratively shifts the test interval toward larger observation parameter values, so that the predicted emission parameter values ​​are advantageously more finely resolved with respect to the observation parameter value and / or are effectively determined toward larger observation parameter values, i.e., particularly toward the future.

[0042] In particular, the test interval is shifted into the future by a predetermined prediction interval length of the second observation parameter interval starting from the old test interval start value if time is selected as the observation parameter. Alternatively, the test interval is shifted, in particular, starting from the old test interval start value by a predetermined prediction interval length of the second observation parameter interval toward a larger, i.e., more distant, distance—and thus effectively also toward the future—if a distance or a route is selected as the observation parameter.

[0043] Particularly preferably, the internal combustion engine is operated in accordance with the predicted emission parameter value after the termination criterion is reached. Alternatively or additionally, the internal combustion engine is operated in accordance with the predicted emission parameter value of the old test interval while the predicted emission parameter value of the new test interval is calculated.

[0044] In particular, at least N predicted emission parameter values ​​are calculated iteratively using equation (3) or (4). In one embodiment, the predicted emission parameter values ​​are determined directly as parameter values. Alternatively, the predicted emission parameter values ​​are determined as functions of the at least one control value, in particular of the plurality of control values, wherein the at least one control value, in particular of the plurality of control values, is varied such that the predicted emission parameter values ​​calculated using the functions each satisfy equations (3) or (4).

[0045] According to a further development of the invention, it is provided that a maximum number of iterations is used as the termination criterion, wherein steps b) to c) are repeated until the maximum number of iterations is reached.

[0046] According to a further development of the invention, time or distance is used as the observation parameter. In particular, the distance is a distance traveled by a vehicle in which the internal combustion engine is operated, in particular for the propulsion of which the internal combustion engine is used, for example a rail vehicle. The distance is thus also, in particular, a distance traveled by the internal combustion engine itself. In particular, based on the application, the legal regulation requires an averaging of the emission parameter value over time or distance. This advantageously makes it possible to monitor and, in particular, comply with the legal regulation both over time and distance.

[0047] According to a further development of the invention, it is provided that the at least one predicted emission parameter value is determined by means of a Gaussian process model.

[0048] In particular, a plurality of predicted emission parameter values, preferably a sequence of predicted emission parameter values ​​successive along the observation parameter, is determined for the second observation parameter interval.

[0049] According to a further development of the invention, the at least one control value, in particular the plurality of control values, for operating the internal combustion engine is obtained based on the at least one predicted emission parameter value. Alternatively or additionally, the at least one control value—in particular the plurality of control values—for operating the internal combustion engine is obtained using the Gaussian process model. Advantageously, the at least one control value—in particular the plurality of control values—is thus determined for, in particular, optimal operation of the internal combustion engine.

[0050] Preferably, an operating plan of the internal combustion engine is created based on the at least one control value - in particular the plurality of control values.

[0051] In a particularly preferred embodiment, the at least one control value for operating the internal combustion engine is determined using a model predictive control algorithm (MPC algorithm) and / or an internal combustion engine model. In particular, a general dynamic optimization problem of the MPC algorithm is solved, taking into account one of equations (1) to (4) as a constraint, and the at least one control value—in particular, the plurality of control values—is thus determined. Advantageously, the internal combustion engine is thus operated in such a way that the statutory emission limit, in particular, is complied with while simultaneously maintaining good engine dynamics.

[0052] According to a further development of the invention, if time is used as the observation parameter, the test interval has a maximum length of 60 seconds. Alternatively or additionally, the predetermined prediction interval length has a maximum length of 500 ms. Alternatively or additionally, the predicted emission parameter values ​​have a time interval of at most 20 ms, preferably at most 10 ms. Alternatively or additionally, the number of iteration steps is selected such that the time interval between the current observation parameter value and the new test interval end value is at most 10 seconds—in particular, the predetermined time prediction horizon is at most 10 seconds.

[0053] In particular, the prediction horizon is chosen such that the slowest system dynamics can be represented and / or have transitioned to a steady state. Alternatively or additionally, the predetermined prediction interval length is chosen such that both the fastest system dynamics can be represented and the computation time is optimized, in particular minimized.

[0054] According to a further development of the invention, at least one parameter selected from a group consisting of a nitrogen oxide concentration, a nitrogen oxide partial pressure, a nitrogen oxide mass, a carbon dioxide concentration, a carbon dioxide partial pressure, a carbon dioxide mass, a hydrocarbon concentration, a hydrocarbon partial pressure, a hydrocarbon mass, a particle concentration, a particle mass, and a combination of the previous parameters is used as the emission parameter.

[0055] In particular, if a combination of the previous parameters is used as emission parameter, a separate equation is set up and solved for each parameter according to equations (1) to (4).

[0056] The invention is explained in more detail below with reference to the drawings, which show: Fig. 1 a schematic representation of an embodiment of an internal combustion engine with a control device for operating the internal combustion engine, Fig. 2 a schematic representation of a first embodiment of a method for operating the internal combustion engine, Fig. 3 a schematic representation of a second embodiment of the method for operating the internal combustion engine, and Fig. 4 a schematic representation of a determination of emission parameter values.

[0057] Fig. 1 shows a schematic representation of an embodiment of an internal combustion engine 1 with a control device 3 for operating the internal combustion engine 1. Optionally, the internal combustion engine 1 additionally has an emissions sensor 5.

[0058] The control device 3 comprises a recording module 7, a computing module 9, and a control module 11. Optionally, the control device 3 additionally comprises an operating plan module 13. In particular, the recording module 7, the computing module 9, and the control module 11 are connected to one another, in particular for data transmission. In particular, the optional operating plan module 13 is also connected to the recording module 7, the computing module 9, and the control module 11, in particular for data transmission.

[0059] The acquisition module 7 is configured to generate a recorded emission parameter value e rof the internal combustion engine 1 over a first observation parameter interval from a previous observation parameter value 15 - in particular x a - X + x p - up to a current observation parameter value x a In particular, the acquisition module 7 is configured to obtain the emission parameter value e r in particular based on a simulation and / or a model, wherein the detection module 7 preferably has a virtual sensor for this purpose or is itself designed as a virtual sensor. Alternatively, the detection module 7 has an interface that is configured to transmit the detected emission parameter value e r to receive, in particular as a measured value of a physical sensor - in particular the emission sensor 5. The recorded emission parameter value e ris then, in particular, a measured emission parameter value. It is also possible for the detection module 7 to be configured to determine, in particular calculate, the emission parameter value e from at least one measured value of a physical sensor.

[0060] The calculation module 9 is configured to calculate at least one predicted emission parameter value e p of the internal combustion engine 1 for a second observation parameter interval from the current observation parameter value x a up to a prediction observation parameter value of 17 - in particular x a + x p - to determine such that an emission parameter value of the internal combustion engine 1 averaged over a test interval 27 comprising the first observation parameter interval and the second observation parameter interval has a predetermined emission parameter limit value e maxIn particular, the calculation module 9 is configured to calculate the at least one predicted emission parameter value e p by means of equation (1) or (2). Alternatively or additionally, the calculation module 9 is particularly configured to calculate the at least one predicted emission parameter value e p directly as a parameter value. Alternatively, the calculation module 9 is particularly configured to determine the at least one predicted emission parameter value e p as a function of at least one control value, in particular a plurality of control values, wherein the calculation module 9 is in particular set up to vary the at least one control value, in particular the plurality of control values, in such a way that the predicted emission parameter value e calculated on the basis of the function p Equation (1) or (2) is satisfied.

[0061] In a preferred embodiment, the calculation module 9 is configured to a) set the previous observation parameter value 15 as an old test interval start value 19 and the prediction observation parameter value 17 as an old test interval end value 21. Furthermore, the calculation module 9 is configured to b) extend the test interval 27 starting from the old test interval start value 19 by a predetermined prediction interval length x p of the second observation parameter interval in the direction of larger observation parameter values ​​x, so that the test interval 27 has a new test interval start value 23 and a new test interval end value 25. In addition, the calculation module 9 is configured to c) at least one new predicted emission parameter value e pof the internal combustion engine 1 for a new second observation parameter interval from the old test interval end value 21 to the new test interval end value 25 such that the emission parameter value of the internal combustion engine 1 averaged over the shifted test interval 27 exceeds the predetermined emission parameter limit value e max does not exceed. In addition, the calculation module 9 is configured to d) repeat steps b) to c) with the new test interval start value 23 as the old test interval start value 19 and the new test interval end value 25 as the old test interval end value 21 until a termination criterion is reached. In particular, the calculation module 9 is configured to calculate at least N predicted emission parameter values ​​e p using equations (3) or (4). Alternatively or additionally, the calculation module 9 is particularly configured to calculate the predicted emission parameter values ​​e pdirectly as parameter values. Alternatively, the calculation module 9 is particularly configured to determine the predicted emission parameter values ​​e p as functions of the at least one control value, in particular of the plurality of control values, wherein the calculation module 9 is in particular set up to vary the at least one control value, in particular the plurality of control values, in such a way that the predicted emission parameter values ​​e calculated on the basis of the functions p satisfy equations (3) or (4) respectively.

[0062] Preferably, the calculation module 9 is configured to use a maximum number of iterations as the termination criterion. The calculation module 9 is configured to repeat steps b) to c) until the maximum number of iterations is reached.

[0063] Particularly preferably, the calculation module 9 is configured to calculate the at least one predicted emission parameter value e p using a Gaussian process model.

[0064] The control module 11 is configured to control the internal combustion engine 1 in accordance with the at least one predicted emission parameter value e p to operate. In particular, the control module 11 is configured to operate the internal combustion engine 1 based on the at least one control value, in particular the plurality of control values.

[0065] In particular, the control device 3 - in particular the computing module 9 and the control module 11 - is configured to use as a control value at least one parameter selected from a group consisting of an introduction time for introducing, in particular jetting or injecting, a fuel, an introduction quantity of the fuel, an introduction pressure, a fresh air mass flow, valve control times for gas exchange valves, in particular intake and exhaust valves, such as in particular a valve opening time and a valve closing time, an internal combustion engine speed and a combination of the previous parameters.

[0066] Particularly preferably, the control device 3 - in particular the control module 11 - is designed to control the internal combustion engine 1 in accordance with the at least one predicted emission parameter value e pafter reaching the termination criterion. Alternatively or additionally, the control device 3 - in particular the control module 11 - is configured to operate the internal combustion engine 1 in accordance with the at least one predicted emission parameter value e p of the old test interval 27, while the calculation module 9 calculates the predicted emission parameter value e p of the new test interval 27 is calculated.

[0067] In particular, the optional operating plan module 13 is configured to, based on the at least one predicted emission parameter value e pto determine the at least one control value, in particular the plurality of control values, for operating the internal combustion engine 1. Alternatively or additionally, the operating plan module 13 is configured to use the Gaussian process model to determine the at least one control value, in particular the plurality of control values, for operating the internal combustion engine 1. Preferably, the operating plan module 13 is configured to create an operating plan for the internal combustion engine 1 based on the at least one control value, in particular the plurality of control values.

[0068] In a particularly preferred embodiment, the operating plan module 13 is configured to determine the at least one control value, in particular the plurality of control values, for operating the internal combustion engine 1 based on a model predictive control algorithm (MPC algorithm) and / or an internal combustion engine model. In particular, the operating plan module 13 is configured to solve a general dynamic optimization problem of the MPC algorithm, taking into account one of equations (1) to (4) as a constraint, and thus to determine the at least one control value, in particular the plurality of control values.

[0069] The optional emission sensor 5 is in particular configured to measure the at least one emission parameter value e r to determine.

[0070] Fig. 2 shows a schematic representation of a first embodiment of a method for operating the internal combustion engine 1, wherein in particular the control device 3 according to Fig. 1 is used to carry out the procedure.

[0071] In a first step S1, a recorded emission parameter value e r of the internal combustion engine 1 over a first observation parameter interval from a previous observation parameter value 15 to a current observation parameter value x a In particular, the emission parameter value e r calculated using a simulation and / or a model. Alternatively, the recorded emission parameter value e r received.

[0072] In a second step S2, at least one predicted emission parameter value e p of the internal combustion engine 1 for a second observation parameter interval from the current observation parameter value xa up to a prediction observation parameter value 17 is determined such that an emission parameter value of the internal combustion engine 1 averaged over a test interval 27 comprising the first observation parameter interval and the second observation parameter interval has a predetermined emission parameter limit value e max In particular, at least one predicted emission parameter value e p calculated using equation (1) or (2). In one embodiment, the at least one predicted emission parameter value e p directly as a parameter value. Alternatively, at least one predicted emission parameter value e pas a function of the at least one control value, in particular the plurality of control values, wherein the at least one control value, in particular the plurality of control values, is varied such that the at least one predicted emission parameter value e calculated on the basis of the function p Equation (1) or (2) is fulfilled. Particularly preferably, the at least one predicted emission parameter value e p determined using a Gaussian process model.

[0073] In a third step S3, the internal combustion engine 1 is controlled in accordance with the at least one predicted emission parameter value e p In particular, based on the at least one predicted emission parameter value e pthe at least one control value, in particular the plurality of control values, for operating the internal combustion engine 1 is obtained. Alternatively or additionally, the at least one control value, in particular the plurality of control values, for operating the internal combustion engine 1 is obtained using the Gaussian process model. Preferably, an operating plan for the internal combustion engine 1 is created based on the at least one control value, in particular the plurality of control values. Particularly preferably, the at least one control value, in particular the plurality of control values, for operating the internal combustion engine 1 is determined using a model predictive control algorithm (MPC algorithm) and / or an internal combustion engine model.In particular, a general dynamic optimization problem of the MPC algorithm is solved taking into account one of the equations (1) to (4) as a constraint and thus the at least one control value, in particular the plurality of control values, is determined.

[0074] In particular, the current observation parameter value x a the observation parameter value x for which the most recent recorded emission parameter value e r Alternatively or additionally, the current observation parameter value x a the instantaneous value of the quantized or discrete observation parameter x in which the calculations are currently being performed.

[0075] In particular, at least one parameter is used as the control value, which is selected from a group consisting of an introduction time for introducing, in particular jetting or injecting, a fuel, an introduction quantity of the fuel, an introduction pressure, a fresh air mass flow, valve control times for gas exchange valves, in particular intake and exhaust valves, such as in particular a valve opening time and a valve closing time, an internal combustion engine speed and a combination of at least two of the previous parameters.

[0076] In particular, the time or the distance is used as the observation parameter x. Preferably, if the time is used as the observation parameter x, the test interval 27 has a time length X of at most 60 seconds. Alternatively or additionally, the predetermined prediction interval length x pa maximum time length of 500 ms. Alternatively, the predicted emission parameter values ​​e p a time interval of no more than 20 ms, preferably no more than 10 ms.

[0077] In particular, at least one parameter selected from a group consisting of a nitrogen oxide concentration, a nitrogen oxide partial pressure, a nitrogen oxide mass, a carbon dioxide concentration, a carbon dioxide partial pressure, a carbon dioxide mass, a hydrocarbon concentration, a hydrocarbon partial pressure, a hydrocarbon mass, a particle concentration, a particle mass, and a combination of the previous parameters is used as the emission parameter e.

[0078] In particular, if a combination of the previous parameters is used as the emission parameter e, a separate equation is set up and solved for each parameter according to equations (1) to (4).

[0079] Fig. 3 shows a schematic representation of a second embodiment of the method for operating the internal combustion engine 1, wherein in particular the control device 3 according to Fig. 1 is used to carry out the procedure.

[0080] In particular, the second embodiment is an extension of the first embodiment according to Fig. 2.

[0081] Identical and functionally equivalent elements are provided with the same reference symbols, so that reference is made to the previous description.

[0082] In particular, the predicted emission parameter values ​​e p calculated using equations (3) or (4). In one embodiment, the predicted emission parameter values ​​e p directly as parameter values. Alternatively, the predicted emission parameter values ​​e pas functions of the at least one control value, in particular of the plurality of control values, wherein the at least one control value, in particular the plurality of control values, are varied such that the predicted emission parameter values ​​e calculated on the basis of the functions p satisfy equations (3) or (4) respectively.

[0083] In particular, equation (3) or (4) is considered, in particular solved, in the second step S2 with n=1.

[0084] In particular, the following steps are carried out between the second step S2 and the third step S3: In a first pass of a fourth step S4, the previous observation parameter value 15 is used as an old test interval start value 19 and the prediction observation parameter value 17 is used as an old test interval end value 21. Furthermore, the test interval 27 is extended starting from the old test interval start value 19 by the predetermined prediction interval length x p of the second observation parameter interval is shifted toward larger observation parameter values ​​x, so that the test interval 27 has a new test interval start value 23 and a new test interval end value 25. In particular, in each further pass of the fourth step S4, the new test interval start value 23 is used as the old test interval start value 19 and the new test interval end value 25 is used as the old test interval end value 21. Based on this, the test interval 27 is then shifted as described above.

[0085] In a fifth step, a new predicted emission parameter value e p of the internal combustion engine 1 for a new second observation parameter interval from the old test interval end value 21 to the new test interval end value 25 is determined such that the emission parameter value of the internal combustion engine 1 averaged over the shifted test interval 27 exceeds the predetermined emission parameter limit value e max In particular, equation (3) or (4) with n=n+1 is considered, in particular solved.

[0086] In a sixth step S6, a termination criterion is checked. In particular, a maximum number of iterations N is used as the termination criterion, with the fourth step S4 and the fifth step S5 being repeated until the maximum number of iterations is reached. In particular, a number of iteration steps N is chosen such that a time interval between the current observation parameter value xa and the new test interval end value 25 is no more than 10 seconds.

[0087] If the termination criterion is not met in the sixth step, the fourth step S4 is performed again, wherein in particular the new test interval start value 23 is used as the old test interval start value 19 and the new test interval end value 25 is used as the old test interval end value 21.

[0088] If the termination criterion is met in the sixth step S6, the third step S3 is carried out. Particularly preferably, the internal combustion engine 1 is operated in accordance with the predicted emission parameter value e p after reaching the termination criterion in the sixth step S6.

[0089] Fig. 4 shows a schematic representation of a determination of emission parameter values ​​e according to the second embodiment of the method from Fig. 3.

[0090] In Fig. 4 is the emission parameter value e - in particular the recorded emission parameter values ​​e r and the predicted emission parameter values ​​e p - represented as a graph in a coordinate system with the axes x - observation parameter - and e - emission parameter value. Furthermore, the test interval 27 is drawn as a rectangle in the coordinate system, wherein the rectangle extends in the direction of the x-axis from the previous observation parameter value 15 to the prediction observation parameter value 17 or from the new test interval start value 23 to the new test interval end value 25. The predetermined emission parameter limit value e max is shown as a horizontal dashed line.

[0091] Fig. 4 a) visualizes a first implementation of the second step S2. In this case, at least one first predicted emission parameter value ep1 of the internal combustion engine 1 for the second observation parameter interval from the current observation parameter value x a up to the prediction observation parameter value 17 is determined such that an emission parameter value of the internal combustion engine 1 averaged over the test interval 27 exceeds the predetermined emission parameter limit value e max does not exceed.

[0092] A first peak of the recorded emission parameter value e r upwards - shown in the test interval 27 on the left - is higher than a second peak of the recorded emission parameter value e r downwards - in the test interval 27 approximately in the middle. Therefore, at least one first predicted emission parameter value ep1 selected below the limit.

[0093] Fig. 4 b) visualizes a second implementation of the second step S2. Here, the previous observation parameter value 15 is used as a first old test interval start value 19' and the prediction observation parameter value 17 is used as a first old test interval end value 21'. In addition, the test interval 27 is extended by the predetermined prediction interval length x starting from the first old test interval start value 19'. p of the second observation parameter interval is shifted towards larger observation parameter values ​​x, so that the test interval 27 has a first new test interval start value 23' and a first new test interval end value 25'. Subsequently, at least one second predicted emission parameter value ep2 of the internal combustion engine 1 for a new second observation parameter interval from the first old test interval end value 21' to the first new test interval end value 25' such that the emission parameter value of the internal combustion engine 1 averaged over the shifted test interval 27 exceeds the predetermined emission parameter limit value e max does not exceed.

[0094] The first peak of the recorded emission parameter value e r upwards is at least partially no longer within the test interval 27, while the second peak downwards is still within the test interval. Therefore, the at least one second predicted emission parameter value ep2 above the limit value because higher emissions can be emitted in the test interval 27.

[0095] Fig. 4 c) visualizes a third implementation of the second step S2. The first new test interval start value 23' is used as a second old test interval start value 19" and the first new test interval end value 25' is used as a second old test interval end value 21". In addition, the test interval 27 is extended starting from the second old test interval start value 19" by the predetermined prediction interval length x p of the second observation parameter interval is shifted towards larger observation parameter values ​​x, so that the test interval 27 has a second new test interval start value 23" and a second new test interval end value 25". Subsequently, at least a third predicted emission parameter value ep3 of the internal combustion engine 1 for a new third observation parameter interval from the second old test interval end value 21" to the second new test interval end value 25" is determined such that the emission parameter value of the internal combustion engine 1 averaged over the shifted test interval 27 exceeds the predetermined emission parameter limit value e max does not exceed.

[0096] The first peak of the recorded emission parameter value e r upwards is no longer completely within the test interval 27, while the second peak downwards is still within the test interval 27. Therefore, at least one third predicted emission parameter value ep3 above the limit value because higher emissions can be emitted in the test interval 27. However, at least one third predicted emission parameter value ep3 below at least one second predicted emission parameter value ep2 since a change in the recorded emission parameter value e r up to the current observation parameter value of Fig. 4 b) after Fig. 4 c) is less than a change of Fig. 4 a) after Fig. 4 b).

Claims

[1] Control device (3) for operating an internal combustion engine (1), with - a detection module (7) configured to detect a detected emission parameter value (e r ) of the internal combustion engine (1) over a first observation parameter interval from a previous observation parameter value (15) to a current observation parameter value (x a ) to obtain, - a calculation module (9) which is arranged to calculate at least one predicted emission parameter value (e p ) of the internal combustion engine (1) for a second observation parameter interval from the current observation parameter value (x a) up to a prediction observation parameter value (17) such that an emission parameter value of the internal combustion engine (1) averaged over a test interval (27) comprising the first observation parameter interval and the second observation parameter interval has a predetermined emission parameter limit value (e max ) and - a control module (11) which is designed to control the internal combustion engine (1) in accordance with the predicted emission parameter value (e p ) to operate. [2] Control device (3) according to claim 1, wherein the computing module (9) is arranged to a) to set the previous observation parameter value (15) as an old test interval start value (19) and the prediction observation parameter value (17) as an old test interval end value (21), and b) the test interval (27) is extended starting from the old test interval start value (19) by a predetermined prediction interval length (x p ) of the second observation parameter interval in the direction of larger observation parameter values ​​(x), so that the test interval (27) has a new test interval start value (23) and a new test interval end value (25), and c) a new predicted emission parameter value (e p ) of the internal combustion engine (1) for a new second observation parameter interval from the old test interval end value (21) to the new test interval end value (25) in such a way that the emission parameter value of the internal combustion engine (1) averaged over the shifted test interval (27) does not exceed the predetermined emission parameter limit value (e max ) and d) repeat steps b) to c) with the new test interval start value (23) as the old test interval start value (19) and the new test interval end value (25) as the old test interval end value (21) until a termination criterion is reached. [3] Control device (3) according to one of the preceding claims, wherein the calculation module (9) is configured to calculate the at least one predicted emission parameter value (e p ) using a Gaussian process model. [4] Control device (3) according to one of the preceding claims, comprising an operating plan module (13) which is configured to, based on the at least one predicted emission parameter value (e p ) and / or to determine at least one control value for operating the internal combustion engine (1) by means of the Gaussian process model and preferably to create an operating plan for the internal combustion engine (1) on the basis of the at least one control value. [5] Control device (3) according to one of the preceding claims, wherein the control module (11) is configured to operate the internal combustion engine (1) based on the at least one control value. [6] Internal combustion engine (1) with a control device (3) according to one of the preceding claims. [7] Internal combustion engine (1) according to claim 6, comprising an emissions sensor (5) which is arranged to measure the at least one detected emissions parameter value (e r ) to determine. [8] Method for operating an internal combustion engine (1), wherein - a recorded emission parameter value (e r ) of the internal combustion engine (1) over a first observation parameter interval from a previous observation parameter value (15) to a current observation parameter value (x a ) is obtained, where - at least one predicted emission parameter value (e p) of the internal combustion engine (1) for a second observation parameter interval from the current observation parameter value (x a ) up to a prediction observation parameter value (17) is determined in such a way that an emission parameter value of the internal combustion engine (1) averaged over a test interval (27) comprising the first observation parameter interval and the second observation parameter interval has a predetermined emission parameter limit value (e max ), whereby - the internal combustion engine (1) in accordance with the predicted emission parameter value (e p ) is operated. [9] Method according to claim 8, wherein a) the previous observation parameter value (15) is used as an old test interval start value (19) and the prediction observation parameter value (17) is used as an old test interval end value (21), wherein b) the test interval (27) is extended starting from the old test interval start value (19) by a predetermined prediction interval length (x p ) of the second observation parameter interval is shifted in the direction of larger observation parameter values ​​(x), so that the test interval (27) has a new test interval start value (23) and a new test interval end value (25), wherein c) a new predicted emission parameter value (e p ) of the internal combustion engine (1) for a new second observation parameter interval from the old test interval end value (21) to the new test interval end value (25) is determined in such a way that the emission parameter value of the internal combustion engine (1) averaged over the shifted test interval (27) does not exceed the predetermined emission parameter limit value (e max ), whereby d) steps b) to c) are repeated with the new test interval start value (23) as the old test interval start value (19) and the new test interval end value (25) as the old test interval end value (21) until a termination criterion is reached. [10] Method according to claim 9, wherein a maximum number of iterations (N) is used as termination criterion, wherein steps b) to c) are repeated until the maximum number of iterations (N) is reached. [11] Method according to one of claims 8 to 10, wherein a time or a distance is used as the observation parameter (x). [12] Method according to one of claims 8 to 11, wherein the at least one predicted emission parameter value (e p ) is determined by means of a Gaussian process model, wherein in particular for the second observation parameter interval a plurality of predicted emission parameter values ​​(e p ) is determined. [13] Method according to one of claims 8 to 12, wherein based on the at least one predicted emission parameter value (e p ) and / or by means of the Gaussian process model at least one control value for operating the internal combustion engine (1) is obtained, wherein an operating plan of the internal combustion engine (1) is preferably created on the basis of the at least one control value. [14] Method according to one of claims 8 to 13, wherein, if time is used as observation parameter (x), - the test interval (27) has a maximum duration of 60 seconds, and / or - the predetermined prediction interval length (x p ) has a maximum duration of 500 ms, and / or - the predicted emission parameter values ​​(e p ) have a time interval of no more than 20 ms, preferably no more than 10 ms, and / or - a number of iteration steps (N) is chosen such that a time interval between the current observation parameter value (x a ) and the new test interval end value (25) is no more than 10 seconds. [15] Method according to one of claims 8 to 14, wherein at least one parameter selected from a group consisting of a nitrogen oxide concentration, a nitrogen oxide partial pressure, a nitrogen oxide mass, a carbon dioxide concentration, a carbon dioxide partial pressure, a carbon dioxide mass, a hydrocarbon concentration, a hydrocarbon partial pressure, a hydrocarbon mass, a particle concentration, a particle mass, and a combination of the previous parameters is used as emission parameter (e).

Images