DE102019214208A1 - Control for an internal combustion engine in a hybrid vehicle, drive train for a hybrid vehicle, hybrid vehicle and method in a control for an internal combustion engine - Google Patents

Abstract

Control for an internal combustion engine in a hybrid vehicle for a correction of a torque determination, which is configured to determine a first approximate torque signal from an actual torque curve of the internal combustion engine, which is established on the basis of a target torque signal, and based on an in / on an electric machine measured EM counter torque signal to correct the first approach torque signal to determine a more accurate second approach torque signal.

Description

The invention relates to a control for an internal combustion engine in a hybrid vehicle for correcting a torque determination, a drive train for a hybrid vehicle, a hybrid vehicle and a method for correcting a torque determination in a control for an internal combustion engine.

It is known that the torque generated or output by an internal combustion engine in a motor vehicle depends on various factors. The main influencing variables are the fuel mass supplied to the combustion chamber, the air-fuel ratio and the engine's frictional torque. The torque output by the internal combustion engine is an important parameter for a large number of functions in drive train control units.

The torque determined in a control for an internal combustion engine can essentially be based on comparatively imprecise calculation methods which are based on the fuel mass supplied per combustion and an estimated value of the friction torque. These are calculated from the signals of several sensors in the intake and exhaust tract and, if necessary, in / on the internal combustion engine. The most important are intake manifold pressure sensors and / or air mass flow sensors and - if available - cylinder pressure sensors.

Furthermore, the torque determined in a control for an internal combustion engine can be determined from a torque characteristic map which was previously measured on a reference internal combustion engine by means of sensors. However, series fluctuations can occur, so that the torques differ from vehicle to vehicle.

Due to sensor tolerances, different components, temperature influences, friction tolerances and comparatively slow dynamic properties of the sensors, the mean torque error is usually in the range of approx. 10%. If the properties of components of the internal combustion engine change (aging, deposits, defects), the error increases significantly.

This often leads to stationary and dynamic discrepancies between the required target torque of the internal combustion engine determined by the driver or the setpoint torque determined in the control unit and the actual torque curve. This can be particularly noticeable in the event of rapid torque changes, such as those that occur when starting the engine, shifting gears or during internal switchovers in the internal combustion engine (e.g. cylinder deactivation).

Torque control methods are known for hybrid or electric vehicles.

From the German Offenlegungsschrift DE 10 2015 115 714 A1 a torque control method for a hybrid vehicle is known. The hybrid electric vehicle can be switched to a specific operating mode. In this operating mode, a torque of an internal combustion engine and two electric machines is determined. The torques are then corrected according to an active anti-slip strategy and a handling stability strategy.

From the German Offenlegungsschrift DE 10 2014 214 541 A1 a drive arrangement for a motor vehicle with a first and a second electric machine is known, the two electric machines being mechanically coupled. If a torque generated by the first electrical machine deviates from a setpoint value, the second electrical machine can compensate for the torque deviation.

In principle, torque sensors and methods for determining them in drive trains or drive arrangements of vehicles are known.

From the German patent specification DE 10 2016 212 113 B3 It is also known to determine the torque transmitted from a first to a second drive element, preferably in electrically operated vehicles or hybrid vehicles, with a strain gauge. The torque is transmitted through a shaft. The two drive elements are connected to one another via a connecting element or a fastening means and the strain measuring element is connected to the two drive elements. Due to the torque, a counter-torque is produced, which leads to a rotation of the two drive elements against each other and thus to a deformation of the connecting element or the fastening means. An evaluation device calculates the transmitted torque from the deformation.

The moreover one is from the German patent specification DE 10 2015 111 409 B3 a method for determining forces and / or torques between two elements that are movable relative to one another or between two areas of an elastically deformable element. The method can be used in drive trains or drive arrangements of vehicles. Both of them are there Movable elements, for example engine output shaft and elastic side shaft to a drive wheel, each have a sensor (transmitter) assigned to them, which output time-discrete and / or value-discrete signals that are synchronized to a common time reference point. Forces and / or torques are calculated from the time profile of the relative movement (relative angle of rotation) and / or a torsion of the elastic side shaft, which is determined from the encoder signals.

The disadvantage of the method described above is the need for additional sensors. A direct measurement of the torque by means of torque sensors is currently not possible in series due to a lack of available installation space and sufficient and inexpensive and small sensors.

The object of the present invention is to provide a correction of a torque determination for an internal combustion engine in a hybrid vehicle which at least partially overcomes the above-mentioned disadvantages.

This object is achieved by the control according to the invention according to claim 1, a drive train according to claim 13, a hybrid vehicle according to claim 14 and a method according to claim 15.

According to a first aspect, the invention provides a control for an internal combustion engine in a hybrid vehicle for a correction of a torque determination, which is set up to
determine a first approximate torque signal from an actual torque curve of the internal combustion engine, which is set based on a target torque signal, and correct the first approximate torque signal based on an EM counter-torque signal measured in / on an electric machine in order to determine a more precise second approximate torque signal.

According to a second aspect, the invention provides a drive train for a hybrid vehicle which has a controller for an internal combustion engine according to the first aspect.

According to a third aspect, the invention provides a hybrid vehicle, the hybrid vehicle having a drive train according to the second aspect.

• bestimmen eines ersten Näherungsdrehmomentsignals von einem tatsächlichen Drehmomentverlauf der Verbrennungskraftmaschine, welcher sich aufgrund eines Soll-Drehmomentsignals einstellt; und
• korrigieren des ersten Näherungsdrehmomentsignals basierend auf einem in/an einer Elektromaschine gemessenen EM-Gegendrehmomentsignals, um ein genaueres zweites Näherungsdrehmomentsignal zu ermitteln.

According to a fourth aspect, the invention provides a method for correcting a torque determination in a controller for an internal combustion engine according to the first aspect, the method comprising:

• determining a first approximate torque signal from an actual torque curve of the internal combustion engine, which is established on the basis of a setpoint torque signal; and
• correcting the first approach torque signal based on an EM counter torque signal measured in / on an electric machine to determine a more accurate second approach torque signal.

Further advantageous embodiments of the invention emerge from the subclaims and the following description of preferred exemplary embodiments of the present invention.

A control according to the invention for an internal combustion engine in a hybrid vehicle for a correction of a torque determination, which is set up to
determine a first approximate torque signal from an actual torque curve of the internal combustion engine, which is set based on a target torque signal, and correct the first approximate torque signal based on an EM counter-torque signal measured in / on an electric machine in order to determine a more precise second approximate torque signal.

The control for the internal combustion engine can basically be an engine control for an internal combustion engine (ECU), which on the one hand receives sensor signals from sensors installed in / on the internal combustion engine (for example crankshaft and camshaft sensors, air mass flow sensors, lambda sensors, intake manifold pressure sensors, temperature sensors, etc.) and on the other receives signals generated by the driver (for example accelerator pedal angle, brake signal, cruise control, etc.) via an interface to a bus system. The controller can also exchange data with other control units installed in the vehicle, such as a control unit for an electric machine (MCU), a drive train control unit (PCU) in a hybrid vehicle or a transmission control unit in a parallel drive train of a hybrid vehicle. The drive train control device can, for example, register and coordinate certain operating situations of the hybrid vehicle.

In general, however, several of the above-mentioned controls and control devices or other controls and control devices installed in the hybrid vehicle can also be combined in one. For example, the control for the internal combustion engine can also be integrated in the drive train control device or combined with the control device for the electric machine or like that. The correction of the torque determination can therefore also take place in such exemplary embodiments in the control for the internal combustion engine, which is then to be understood, for example, as a component or functional unit in the higher-level control unit. An exemplary embodiment as a separate control for the internal combustion engine is therefore not to be understood as limiting.

In addition, the control for the internal combustion engine can keep operationally relevant programs and data, maps (torque map or torque changes map) and the like in a memory. Correction values for various operating and control variables, which can change from time to time, because, for example, the control has carried out a correction process, can also be kept in the memory.

From the stored data and the signals received (here operating variables), the control for the internal combustion engine can determine other control variables that are relevant for setting the operating range (e.g. speed and requested target torque signal) of the internal combustion engine (e.g. supplied fuel mass, requested target torque, etc. ) and determine the necessary manipulated variables for the actuators installed in the internal combustion engine and control them accordingly.

In order to be able to carry out the above-mentioned functions, the controller has electronic components, storage units and circuits and the like. For example, the controller can contain one or more processors (CPUs), one or more microcontrollers, FPGAs or the like. The storage units can be, for example, magnetic memories or semiconductor memories. The interface to the bus system can be, for example, a CAN interface in order to communicate with a CAN bus system. The bus system can also be a LIN or FlexRay or the like.

The internal combustion engine can in principle be any internal combustion engine, e.g. Otto engine, diesel engine, etc.

The hybrid vehicle can be a vehicle with a parallel or serial drive train, the hybrid vehicle having an internal combustion engine and at least one electric machine. A hybrid vehicle with a parallel drive train also has a transmission. The electric machine can be a synchronous or an asynchronous electric machine.

The torque that is applied to a transmission in a hybrid vehicle with a parallel drive train, hereinafter referred to as the transmission input torque, is the sum of the actual torque of the internal combustion engine and an (actual) EM torque of the electric machine.

As mentioned above, a determination of an actual torque curve of an internal combustion engine, which is not known due to a lack of sensors, is often imprecise. It is therefore desirable to correct the torque determination in order to obtain a more accurate value.

The controller determines a first approach torque signal from the actual torque curve. As mentioned at the beginning, this can be determined, for example, on the basis of the supplied fuel mass and an estimate of the friction torque or it can be determined from a torque map for a respective operating range (e.g. speed and requested target torque signal) of the internal combustion engine.

The first approach torque signal can be an absolute torque or a change in torque. The first approximate torque signal can, for example, be a single value at a specific point in time and / or in a specific operating situation (e.g. vehicle standstill, braking, etc.) and / or in a specific operating range of the actual torque curve. The first approximate torque signal can be a time sampling of the actual torque curve consisting of several values.

The actual torque curve is established on the basis of a target torque signal which, as mentioned above, is determined from various operating variables in the control for the internal combustion engine and which results in the requested target torque signal.

An independent test torque signal can also be applied to the requested target torque signal in order to generate the target torque signal.

The accuracy of the first approximate torque signal can now be improved in that the electric machine present in the drive train of the hybrid vehicle is used to measure an absolute torque or a change in torque. The reason for this is, among other things, that a determination of an EM counter-torque signal on the basis of the more precise sensors installed in / on an electric machine is much more precisely possible. The control unit of the electric machine in the drive train determines the EM counter-torque signal from the sensor signals generated in / on the electric machine, which corresponds very precisely to the (actual) EM torque or the (actual) EM torque change.

To determine the EM counter-torque signal, the actual torque curve generated by the internal combustion engine is compensated for by an equally large counter-torque on the electric motor in suitable operating situations. As mentioned above, this counter torque can be determined comparatively precisely in the control unit of the electric machine.

For example, to carry out the correction, the drive train control unit can determine from various sensor signals (e.g. vehicle speed, pedal angle, speed, brake signal, open clutch, etc.) whether the transmission input torque is zero or remains constant. The clutch is opened, for example, by the transmission control unit and can be precisely recorded.

For example, if the transmission input torque is zero (idling, coasting with the clutch open, vehicle standstill) because the clutch is open, this can be clearly detected and the EM torque can then be set so that it just compensates for the actual torque curve. The EM counter-torque signal thus obtained can be used to determine a correction for the first approach torque signal in order to determine a more accurate second approach torque signal.

In general, however, the necessary conditions for performing the correction can be narrowed down from various sensor signals.

In the event of an applied deviation (test torque signal) to the internal combustion engine, the electric machine can be set in such a way that it just generates the counter-torque of the deviation in order to keep the transmission input torque constant. The counter torque of the electric machine can be measured very precisely, so that the EM counter torque signal is obtained for the correction.

The control unit of the electric machine can forward the EM counter-torque signal to the control of the internal combustion engine. Therefore, based on an EM counter-torque signal measured in / on an electrical machine, the first approach torque signal is corrected in order to determine a more precise second approach torque signal.

The more precise second approximation torque signal can then be used to determine the manipulated variables of the internal combustion engine. In addition, the more precise second approximate torque signal can be stored in the control for each operating range of the internal combustion engine and a corrected, more precise torque characteristic map and / or torque changes characteristic map can thus be obtained. In addition, the discrepancy between the first approach torque signal and the EM torque signal can be entered in a torque correction map, so that the second more accurate approach torque signal is then determined during driving as the sum of the first approach torque signal and the correction value from the torque correction map . This can be done equally for absolute torques (torque correction map) and torque changes (torque changes correction map).

Characteristic maps are a table-like, simple type of model that is not very demanding in terms of computer resources and depicts the relationship between input and output variables of a system. That is why they are used in control units, especially for adaptation maps. In the case of adaptation maps, the value (content) can be changed during operation on the basis of measurement or calculation results (adaptation result).

Almost any mathematical relationship or formula can be represented with characteristic diagrams, whereby the number of input variables is limited. At present, characteristic diagrams with up to four input variables are known in control units.

Further methods for storing adaptation results in controls and control units can be, for example, physical models, polynomial models, neural networks or LOLIMOT (Local Linear Model Trees), which are suitable as an alternative to the use of characteristic maps. These require less storage space and can process a higher number of input variables, but these require more computational effort. At the moment, these methods are rarely used, as parameterization (for example training a neural network) requires a lot of time and computational effort.

In processes with artificial intelligence, neural networks are currently emerging and are finding their way into vehicle control units in control units, for example for automated driving or voice input.

In principle, these methods are also suitable as an alternative to torque or calorific value maps or their correction maps and could be used in vehicles with increasing computing power.

Likewise, the adaptation results (correction values) could, for example, also be transmitted to a cloud via a network connection (for example a wireless network connection) and stored there (for example in a memory on a server). In such exemplary embodiments, the correction value for a specific operating range can be requested via the network connection, so that the second more precise approach torque signal is then determined during driving operation as the sum of the first approach torque signal and the correction value from the cloud.

Furthermore, the above-mentioned processes could also be outsourced to the cloud and no longer calculated in a control device, but in the cloud. This is also conceivable for the torque determination described herein in the control for the internal combustion engine, which is then carried out entirely or partially in the cloud.

The more precise determination of the actual torque curve of the internal combustion engine with the aid of the electric machine present in the hybrid vehicle is advantageous since no additional sensors or components have to be installed in the internal combustion engine and is therefore cost-effective. In addition, this allows the fuel mass and / or the air-fuel ratio to be set more precisely in an operating range. This is therefore also relevant e.g. for exhaust aftertreatment.

Furthermore, in hybrid vehicles with parallel drive trains that have a transmission, the precision of the interaction between the internal combustion engine, electric machine and transmission can be improved. This can have a positive effect, e.g. when starting the internal combustion engine, when shifting gears or load changes, or in general in dynamic operating situations.

In some exemplary embodiments, the controller is set up to apply a test torque signal to a requested target torque signal in order to generate the target torque signal.

The test torque signal can, for example, be an additional torque (increase or decrease) for a short period of time, which can cause a small change in the actual torque curve. However, it can also consist of several additional torques connected one behind the other.

The manipulated variables of the internal combustion engine are set in accordance with the target torque signal and thus deviate from the requested target torque signal when the test torque signal is switched on. Accordingly, the internal combustion engine reacts to the test torque signal. The amplitude of the test torque signal is small compared to the requested target torque signal.

If the test torque signal is an additional torque (increase or decrease) for a short period of time, then in such exemplary embodiments, a first approximate torque signal from the actual torque curve in an operating range of the internal combustion engine is an estimate of a change in the actual torque curve. As described above, the change is compensated for by the electric machine and measured comparatively precisely, so that an EM counter-torque signal can be used to correct the first approximate torque signal. Therefore, in such exemplary embodiments, a change in torque can be determined more precisely in an operating range.

Deviations between the first approximate torque signal and the EM counter-torque signal indicate increased tolerances, aging or faults or defects in the internal combustion engine, e.g. incorrect fuel metering (e.g. the injectors, etc.) or incorrect air mass flow (e.g. due to aged sensor, leaky intake manifold, defective Turbocharger, etc.).

In some exemplary embodiments, the test torque signal consists of several test torque pulses.

The test torque signal can represent, for example, a binary pseudo-random signal (PRBS) with a predetermined number of test torque pulses and thus a predetermined duration, with the test torque pulses having a low amplitude. The control for the internal combustion engine can have a test pattern generator for this purpose.

In such exemplary embodiments, a value of the first approximate torque signal and a value of the EM counter-torque signal can be determined for each new test torque pulse (torque changes in the actual torque curve), which compensates the test torque pulse as best as possible. In the case of a correlation of the total values at the end of the Test torque signal, any value of the first approximation torque signal can be corrected based on the EM counter torque signal.

Correlation, for example cross-correlation, is a statistical method for determining a statistical relationship between two signals. One advantage of the correlation is that the correlation is very sensitive and, at the same time, less susceptible to interference. As a result, even small changes in torque in the actual torque curve can be detected in normal driving or dynamic operating situations.

In some exemplary embodiments, the controller is configured to correct the EM counter-torque signal based on the requested target torque signal in order to obtain a corrected EM counter-torque signal.

During normal driving of a hybrid vehicle, the requested target torque signal changes over time. In order to keep the influence of this change on the determination of the correlation low, for example, the requested setpoint torque can be used to correct the EM counter-torque signal. For example, in the event of an acceleration of the hybrid vehicle, the increase in the requested target torque can be calculated accordingly from the EM counter-torque signal in order to obtain a corrected EM counter-torque signal.

In some exemplary embodiments, the controller further has at least one shift register, and the controller is set up to have several values of an input signal, which is the sum of the requested target torque signal and the first approximate torque signal, and several values of an output signal which corresponds to the corrected EM counter-torque signal, at times within the duration of the test torque signal in the shift register.

The shift register or registers is, for example, a logic switching mechanism which can consist of memory cells connected in series, which can preferably store whole numerical values, and which can work according to the FIFO principle (first in first out).

The manipulated variables for setting the internal combustion engine are set in accordance with the target torque signal. If, for example, a test torque signal from several test torque pulses is applied to the requested target torque, the changes in the actual torque curve caused by the test torque pulses can be estimated as the first approximate torque signal. Therefore, the values of the sum of the first approximate torque signal and the requested target torque determined for each test torque pulse form the input signal for determining a correlation with the output signal, which in such exemplary embodiments is the corrected EM counter-torque signal.

The shift register is preferably designed in such a way that all values can be stored up to the end of the test torque signal.

In some exemplary embodiments, the times of the values of the input and output signals differ by a predetermined time delay.

In general, it is to be expected that the internal combustion engine will implement the test torque pulses applied with a time delay. It is therefore advisable in such situations to determine the values of the output signal (corrected EM counter-torque signal) also with a time delay in order to achieve the greatest possible correlation between the input and output signals.

In some exemplary embodiments, the controller is set up to determine a correlation coefficient from the stored values of the input and output signals.

The correlation coefficient ( CCC ) is a statistical variable and a measure for the similarity between two signals or for a statistical relationship between two signals. For example, there is a correlation coefficient of 1 for the same signal curves and a direct relationship between the two signals. A value of 0, on the other hand, means that there is no similarity between the signals.

In some exemplary embodiments, the time delay is selected in accordance with a maximum value of the correlation coefficient.

As a result, the relationship between the two signals is greatest and a correction of the first approach torque signal becomes possible and more accurate. The time delay can be determined experimentally for each operating range.

In some exemplary embodiments, the controller is set up to determine a weighting factor from the stored values of the input and output signals.

In some exemplary embodiments, the controller is configured to generate the more precise second approximation torque signal based on the to determine stored values of the output signal and the weighting factor.

In some exemplary embodiments, the controller is set up to determine the more precise second approximation torque signal only if the correlation coefficient is above a predetermined threshold value.

Correction of the first approximation torque signal based on the stored values of the output signal and the weighting factor only makes sense if the input and output signals have a certain degree of similarity.

In some exemplary embodiments, the controller is set up to determine the more precise second approximation torque signal only when one or more release conditions are met.

The release conditions can be: maximum torque of the electric machine greater than or equal to the maximum torque of the internal combustion engine, speed in the permissible range, load in the permissible range, speed and load dynamics, ambient pressure below the limit value, ambient temperature above the limit value, cooling water temperature above the limit value, on-board voltage in the permissible range , Diagnoses of all involved sensors and actuators carried out completely and no errors detected, waiting time for repeating the correction at the current operating point has expired and transmission input torque constant or changeable. Instead of a time, a selectable number of starts of the internal combustion engine or a distance traveled can also be used, for example.

When correcting absolute torques, the transmission input torque must also be zero or known.

Furthermore, the implementation of the correction torque determination can be aborted if the release conditions are no longer met, since excessive inaccuracy of the result is to be expected. As a result, the above conditions must also be checked continuously while the correction torque determination is active.

Some exemplary embodiments relate to a drive train for a hybrid vehicle which has a controller for an internal combustion engine as described herein.

Some exemplary embodiments relate to a hybrid vehicle, the hybrid vehicle having a drive train as described herein.

• bestimmen eines ersten Näherungsdrehmomentsignals von einem tatsächlichen Drehmomentverlauf der Verbrennungskraftmaschine, welcher sich aufgrund eines Soll-Drehmomentsignals einstellt; und
• korrigieren des ersten Näherungsdrehmomentsignals basierend auf einem in/an einer Elektromaschine gemessenen EM-Gegendrehmomentsignals, um ein genaueres zweites Näherungsdrehmomentsignal zu ermitteln.

Some exemplary embodiments relate to a method for correcting a torque determination in a controller for an internal combustion engine, the method comprising:

• determining a first approximate torque signal from an actual torque curve of the internal combustion engine, which is established on the basis of a setpoint torque signal; and
• correcting the first approach torque signal based on an EM counter torque signal measured in / on an electric machine to determine a more accurate second approach torque signal.

• 1 zeigt schematisch einen Aufbau eines parallelen Antriebsstrangs in P1-Anordung 1 in einem Hybrid-Fahrzeug.

Embodiments of the invention will now be described by way of example and with reference to the accompanying drawing, in which:

• 1 shows schematically a structure of a parallel drive train in P1 arrangement 1 in a hybrid vehicle.

The parallel drive train in P1 arrangement 1 has an internal combustion engine 2 on that on a crankshaft 3 an actual torque curve Mv generated. An electric machine 4th is present in parallel and generates an EM torque M _EM , which is in addition to the actual torque curve Mv works. A clutch 5 is located between the electric machine 4th and a gearbox 6th . A transmission input shaft 7th transfers the sum of the actual torque curve M _v the internal combustion engine 2 and EM torque M _EM to the gearbox so that the gearbox input torque M _TMin stationary is given by: $${\mathrm{M.}}_{T\mathrm{M.}in}={\mathrm{M.}}_V+{\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}}.$$

A torque is applied to one or more wheels 8th transmitted. A powertrain control unit 9 is via a bus system, here a CAN bus system, with a control for the internal combustion engine 10 and the control unit for the electric machine 11 connected. The control unit for the electric machine 11 controls the electric machine 4th through power electronics 12th at.

In this embodiment, the powertrain control device 9 , the control for the internal combustion engine 10 and the control unit for the electric machine 11 designed as separate controls or control units. However, as mentioned above, this is in no way to be understood as limiting.

There is a capture of the EM torque M _EM by the control unit of the electric machine 11 from the electrical currents, voltages and a control frequency (by the power electronics 12th ) can be made much more precise using inexpensive, comparatively accurate current sensors, the actual torque curve can M _v can be easily determined from the above equation when the transmission input torque M _TMin is known.

However, for reasons of cost and space, there is no sensor for in the parallel drive train in P1 arrangement 1 M _Tmin installed. The absolute torque in an operating range of the internal combustion engine 2 is then determined in suitable operating situations in which the wheel torque requested by the driver is equal to zero (thrust, braking, sailing with the clutch open, vehicle standstill, easy downhill travel, etc.), since the transmission input torque is then also stationary M _TMin equals zero. The operating situation is controlled by the drive train control unit 9 recognized and checked, i.e. the determination is aborted as soon as the operating situation is no longer given (release conditions). To do this, the powertrain control unit sends 9 a release signal to the control for the internal combustion engine 10 .

The release conditions in this exemplary embodiment are: Maximum torque of the electric machine 4th greater than or equal to the maximum torque of the internal combustion engine 2 , Speed in the permissible range, load in the permissible range, ambient pressure below the limit value, ambient temperature above the limit value, cooling water temperature above the limit value, on-board electrical system voltage in the permissible range, diagnoses of all sensors and actuators involved completely carried out and no errors detected, waiting time for repeating the correction in the current operating range has expired and transmission input torque zero or known.

The control for the internal combustion engine 10 then determines a first approach torque signal M _VEi of the actual torque curve Mv .

The EM torque M _EM is set in such a way that the counter-torque to the internal combustion engine is generated precisely. That in / on the electric machine 4th measured EM counter torque signal M _EMi (calculated in the control unit of the electric machine 12th ) is then used to correct the torque determination in the control for the internal combustion engine 10 used. An absolute torque deviation ΔM _VEid between the first approach torque signal M _VEi and the EM counter torque signal M _EMi is given by (in the following only the amounts of the sizes are taken into account): $$\Delta {\mathrm{M.}}_{V\mathrm{E.}id}\left(k\right)={\mathrm{M.}}_{V\mathrm{E.}i}\left(k\right)-{\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}i}\left(k\right).$$

The index k indicates that this deviation is specific to the operating range of the internal combustion engine.

Correcting the first approach torque signal M _VEi then gives a more accurate second approach torque signal M _VEik which is calculated as follows: $${\mathrm{M.}}_{V\mathrm{E.}ik}\left(k\right)={\mathrm{M.}}_{V\mathrm{E.}i}\left(k\right)-\Delta {\mathrm{M.}}_{V\mathrm{E.}id}\left(k\right).$$

In this exemplary embodiment, the detected deviations ΔM _VEid (k) successively in different areas of operation with the index k are determined and entered in a torque correction map. The correction values for different operating ranges with index k must be determined one after the other. Depending on the accuracy requirements, it may also be necessary for an operating point k carry out several measurements in succession and average the results.

2 shows schematically a structure of a parallel drive train in P2 arrangement 20 in a hybrid vehicle.

In the parallel drive train in P2 arrangement, there is only the sequence of clutch 5 and electric machine 4th reversed. The above are therefore also valid in this exemplary embodiment.

In this embodiment, the powertrain control device 9 , the control for the internal combustion engine 10 and the control unit for the electric machine 11 designed as separate controls or control units. However, as mentioned above, this is in no way to be understood as limiting.

3 shows schematically a structure of a serial drive train 30th in a hybrid vehicle

The serial drive train 30th has no gear 6th on. The internal combustion engine 2 In this arrangement it is used to charge a traction battery 13th via the electric machine 4th in generator mode. The hybrid vehicle is driven by a second electric machine 14th by the traction battery 13th is supplied with energy.

With the serial drive train 30th is the transmission input torque M _TMin stationary always equal to zero, since there is no gearbox 6th is available. Thus, in the serial drive train 30th always determine the second approach torque signal as the absolute torque according to the above considerations. The powertrain control unit 9 thus always sends a release signal to the Control for the internal combustion engine 10 as long as the following release conditions are met: Maximum torque of the electric machine 4th greater than or equal to the maximum torque of the internal combustion engine 2 , Speed in the permissible range, load in the permissible range, ambient pressure below the limit value, ambient temperature above the limit value, cooling water temperature above the limit value, on-board electrical system voltage in the permissible range, diagnoses of all sensors and actuators involved completely carried out and no errors detected, and waiting time for repeating the correction in the current operating range expired.

In this embodiment, the powertrain control device 9 , the control for the internal combustion engine 10 and the control unit for the electric machine 11 designed as separate controls or control units. However, as mentioned above, this is in no way to be understood as limiting.

4th shows an exemplary embodiment for determining a correction of a change in torque of an internal combustion engine 2 .

In the following, for the in 4th illustrated method a parallel drive train 1 and 20th of the embodiments of 1 and 2 provided. That too 1 and 2 described method for determining a correction of an absolute torque of the internal combustion engine 2 using the parallel drivetrain in a hybrid vehicle 1 and 20th built-in electric machine 4th sets a known transmission input torque M _TMin ahead. Therefore, this method can usually not be used while driving, because M _TMin cannot be determined while driving.

This in 4th However, the method shown enables a correction for a change in torque of the internal combustion engine to be determined 2 . In the 4th The operating situation shown corresponds to a speed-controlled drive on a level road, so that the transmission input torque should be constant over the period of the correction process.

At 41, a test torque signal, which corresponds to a short-term increase in the setpoint torque signal, is applied to the requested setpoint torque. This leads to a change in an actual torque curve ΔMv (which would otherwise not arise due to the operating situation assumed here), which occurs at 42.

At 42, therefore, the drive train control unit 9 recognized that the change in the actual torque curve ΔM _v has set and reports to the control unit of the electric machine 11 the command to change the EM torque ΔM _EM in the electric machine 4th to generate which is the change in the actual torque curve ΔM _v just compensated for the transmission input torque M _TMin keep constant. This occurs at 43, so that the transmission input torque M _TMin assumes the original value or the original operating situation is established. Because stationary should apply: $${\mathrm{M.}}_{T\mathrm{M.}m}=kOnst.={\mathrm{M.}}_v+{\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}}+\Delta {\mathrm{M.}}_v-\Delta {\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}}$$

A first approach torque signal is then output between 43 and 44 ΔM _VEi , which in this embodiment is a change in torque, in the control for the internal combustion engine 10 and an EM counter torque signal ΔM _EMi in the control unit of the electric machine 11 certainly. Then there will be a deviation between the torque changes ΔΔM _VEid of the first approach torque signal ΔM _VEi and the EM counter torque signal ΔM _EMi certainly: $$\Delta \Delta {\mathrm{M.}}_{V\mathrm{E.}id}\left(k\right)=\Delta {\mathrm{M.}}_{V\mathrm{E.}i}\left(k\right)-\Delta {\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}i}\left(k\right).$$

The index k indicates that this deviation is specific to the operating range of the internal combustion engine.

Correcting the first approach torque signal ΔM _VEi then gives a more accurate second approach torque signal ΔM _VEik which is calculated as follows: $$\Delta {\mathrm{M.}}_{V\mathrm{E.}ik}\left(k\right)=\Delta {\mathrm{M.}}_{V\mathrm{E.}i}\left(k\right)-\Delta \Delta {\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}id}\left(k\right).$$

In this embodiment, the detected deviations ΔΔM _VEid (k) are successively in different operating ranges with the index k are determined and entered in a torque changes correction map. The correction values for different operating ranges with index k must be determined one after the other. Depending on the accuracy requirements, it may also be necessary for an operating point k carry out several measurements in succession and average the results.

From the control for the internal combustion engine 10 is at an operating point with index k a first approximation torque signal ΔM _VEi (k) = 5 Nm is determined. The change in the actual torque curve of the internal combustion engine 2 however is only ΔM _v = 4 Nm (because too little additional fuel was injected). The EM counter torque _signal is also ΔM EMi (k) = 4 Nm. This means that the correction value ΔΔM _VEid (k) = 1 Nm and the more precise second approximation torque _signal is ΔM VEik (k) = 4 Nm.

The release conditions in this exemplary embodiment are: Maximum torque of the electric machine 4th greater than or equal to the maximum torque of the internal combustion engine 2 , Speed in the permissible range, load in the permissible range, speed and load dynamics in the permissible range, ambient pressure below the limit value, ambient temperature above the limit value, cooling water temperature above the limit value, on-board electrical system voltage within the permissible range, diagnoses of all sensors and actuators involved completely carried out and no errors detected, waiting time for The correction has been repeated in the current operating range and the transmission input torque is constant.

The method is advantageous because it can be carried out while driving (with a constant operating situation) and no influencing of driving behavior, smoothness of the internal combustion engine or emissions are to be expected, since the required amplitudes of the change in the actual torque curve ΔM _v are so small that they cannot be felt.

5 shows an embodiment of a control for an internal combustion engine 10 in a hybrid vehicle with a parallel drive train 1 and 20th .

That too 4th described method for determining a correction of a change in torque of the internal combustion engine 2 using the parallel drivetrain in a hybrid vehicle 1 and 20th built-in electric machine 4th sets a constant transmission input torque M _TMin ahead. In real operating situations, this is very seldom or often not the case for a long enough time. It is possible, however, to determine changes in the actual torque curve more precisely while driving with a changing actual torque curve or a changing requested target torque signal by means of the method referred to below as the correlation method.

In this embodiment, the controls are for the internal combustion engine 10 and the control unit for the electric machine 11 designed as separate controls or control units. However, as mentioned above, this is in no way to be understood as limiting.

The release conditions for the correlation method are: Maximum torque of the electric machine 4th greater than or equal to the maximum torque of the internal combustion engine 2 , Speed in the permissible range, load in the permissible range, speed and load dynamics in the permissible range, ambient pressure below the limit value, ambient temperature above the limit value, cooling water temperature above the limit value, on-board electrical system voltage in the permissible range, diagnoses of all sensors and actuators involved completely carried out and no errors detected, and waiting time expired to repeat the correction in the current operating range.

The control for the internal combustion engine 10 has a processor with integrated memory 50 who coordinates and partially takes over the basic control tasks described above. The processor with integrated memory 50 can basically access all other components described below and exchange data and signals. The processor with integrated memory 50 activates the correlation procedure when the release conditions are met.

The subdivision of the control for the internal combustion engine into components only serves to illustrate and describe the mode of operation and is in no way to be understood as limiting.

The control for the internal combustion engine 10 has a target calculation component 51 on which a requested target torque signal M _VEs determined. A test pattern generator 52 generates a test torque signal, in this embodiment a PRBS, consist of test torque pulses ΔM _VEsi (i = 1, ..., N), which corresponds to the requested target torque signal M _VEs are switched on in order to generate a target torque signal. The manipulated variable component is determined from the target torque signal 53 the manipulated variables for controlling the actuators for the internal combustion engine 2 set accordingly. The manipulated variables are sent via a CAN interface 54 to a CAN bus system 55 to the internal combustion engine 2 transmitted.

An actual torque curve is then produced Mv on. The actual torque curve Mv is influenced by the test torque pulse ΔM _VEsi and a first approach torque signal ΔM _VEi is determined. The sum of the requested target torque signal M _VEs and the first approach torque signal ΔM _VEi forms an input signal x _i (i = 1, ..., N): $$x_i={\mathrm{M.}}_{V\mathrm{E.}s}+\Delta {\mathrm{M.}}_{V\mathrm{E.}i}.$$

The input signal x _i is used for each test torque pulse ΔM _VEsi in a shift register 56 saved.

The electric machine 4th is now set so that the change is caused by the test torque pulse ΔM _VEsi in the actual torque curve Mv is compensated as best as possible. As the variable requested target torque signal M _VEs (dynamic operating situation) how a disturbance acts can, under certain circumstances, be reduced by the test torque pulse ΔM _VEsi caused change in the actual torque curve Mv do not compensate exactly as in the above exemplary embodiments. The setting is therefore delayed (time delay τ ) to the test torque pulse ΔM _VEsi . The time delay τ is determined experimentally to have a high statistical relationship between the input signal x _i and the compensation by the electric machine 4th as described below.

An EM counter torque signal ΔM _EMi is measured in / on the electric machine and sent to a dynamic correction component via the CAN bus system 57 transmitted. The dynamic correction component 57 corrects the EM counter torque signal ΔM _EMi based on the requested target torque signal M _VEs by adding a dynamic correction value a _dki from the EM counter torque signal ΔM _EMi is subtracted. This allows systematic influences to be reduced. This gives a corrected EM counter torque signal and forms an output signal y _j (j = 1, ... N): $$y_i=\Delta {\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}i}-a_{dki}.$$

The output signal y _j is used for each test torque pulse ΔM _VEsi in a shift register 58 saved.

A correlator 59 accesses the input signal x _i and the output signal y _j in the shift registers 56 and 58 to get a correlation coefficient CCC and a weighting factor b _yx to calculate. The correlation coefficient CCC lies between 0 and 1 and is a measure of the similarity or correspondence of the input signal x _i and the output signal y _j . A correlation coefficient of 1 results for the same signal curves and means a direct relationship between the two signals. A value of 0, on the other hand, means that there is no similarity between the signals.

$$s_{yy}={\displaystyle \sum {\left(y_j-AM\left(y_j\right)\right)}^2,}$$

$$s_{xy}={\displaystyle \sum \left(x_i-AM\left(x_i\right)\right)\cdot \left(y_j-AM\left(y_j\right)\right)},$$

$$CCC=\frac{{{\displaystyle s}}_{xy}^2}{s_{xx}\cdot s_{yy},}$$

$$b_{yx}=\frac{s_{xy}}{{{\displaystyle s}}_{yy}^2}.$$

The correlation coefficient CCC and the weighting factor b _yx are calculated as follows: $$s_{xx}={\displaystyle \sum {\left(x_i-\mathrm{A.}\mathrm{M.}\left(x_i\right)\right)}^2,}$$

$$s_{yy}={\displaystyle \sum {\left(y_j-\mathrm{A.}\mathrm{M.}\left(y_j\right)\right)}^2,}$$

$$s_{xy}={\displaystyle \sum \left(x_i-\mathrm{A.}\mathrm{M.}\left(x_i\right)\right)\cdot \left(y_j-\mathrm{A.}\mathrm{M.}\left(y_j\right)\right)},$$

$$\mathrm{C.}\mathrm{C.}\mathrm{C.}=\frac{{{\displaystyle s}}_{xy}^2}{s_{xx}\cdot s_{yy},}$$

$$b_{yx}=\frac{s_{xy}}{{{\displaystyle s}}_{yy}^2}.$$

Here are AM ( x _i ) and on( y _j ) the arithmetic mean values of the input signal x _i and the output signal y _j . Exceeds the correlation coefficient CCC a certain threshold CCCs , both signals have such a high degree of correspondence that a determination of the more precise second approximation torque signal ΔM _VEik is possible. The time delay τ is set according to a maximum correlation coefficient. The more accurate second approach torque signal ΔM _VEik is then calculated as follows: $$\Delta {\mathrm{M.}}_{V\mathrm{E.}ik}\left(k\right)=b_{yx}\left(k\right)\cdot f_{mk}\left(k\right)\cdot \Delta {\mathrm{M.}}_{\mathrm{E.}\mathrm{M.}i}\left(k\right).$$

The factor f _mk (k) is a correction factor map with the amplitude deviations between the more precise second approximation torque signal ΔM _VEik and the EM counter torque signal ΔM _EMi be balanced due to systematic influences. In the best case, f _mk (k) is 1 everywhere.

In this exemplary embodiment, the more precise second approximate torque _signals .DELTA.M VEik (k) are successively indexed in different operating ranges k determined and in a torque changes correction map 60 registered. The correction values for different operating ranges with index k must be determined one after the other. Depending on the accuracy requirements, it may also be necessary for an operating point k carry out several measurements in succession and average the results.

The correlation method is advantageous because it can be carried out while driving and no influencing of driving behavior, smoothness of the internal combustion engine or emissions are to be expected. In the correlation method described, the required amplitudes of the test torque pulses are ΔM _VEsi (i = 1, ..., N) so small that they cannot be felt. With the correlation method, even small changes in torque can be seen in the actual torque curve M _v can also be demonstrated in normal driving or dynamic operating situations.

6th shows a principle for a correction of an EM counter-torque signal ΔM _EMi with a dynamic correction value a _dki for the correlation method.

The operating situation in 6th is an accelerating hybrid vehicle in which the correlation method is used. The requested target torque signal increases accordingly as a result of the acceleration. As the variable requested target torque signal M _VEs (dynamic operating situation) how a disturbance acts can, under certain circumstances, be reduced by the test torque pulse ΔM _VEsi caused change in the actual torque curve M _v do not compensate exactly as in the above exemplary embodiments. Therefore, the EM counter torque signal also rises ΔM _EMi systematically.

The EM counter-torque signal is used to reduce the systematic influence on the determination of the correlation and thus to increase the accuracy of the method ΔM _EMi with dynamic correction value a _dki corrected to obtain a corrected EM counter torque signal which is the output signal y _j forms. Thus are in the output signal y _j only the proportions of the test torque pulse ΔM _VEsi caused change in the actual torque curve Mv in the EM counter torque signal ΔM _EMi contain. This becomes the correlation coefficient CCC greater.

7th shows a flowchart for a correction of a torque determination in a control for an internal combustion engine.

The method refers to the exemplary embodiments according to 1 and 2 based. This is in no way to be understood as limiting.

At 65, a first approach torque signal is received M _VEi from an actual torque curve Mv the internal combustion engine 2 determines which occurs on the basis of a target torque signal, as described above.

At 66 becomes the first approach torque signal M _VEi based on an EM counter-torque signal measured in / on an electric machine M _EMi corrected to provide a more accurate second approach torque signal M _VEik to be determined as described above.

8th shows a flow diagram for the implementation of the correlation method.

The method refers to the exemplary embodiment according to 5 based.

At 70 the internal combustion engine is started and initialized.

The correlation method is started at 71 if the release conditions (as described above) are met.

At 72, a test torque pulse is applied ΔM _VEsi to the requested target torque signal M _VEs activated as described above.

At 73, the first approach torque signals become ΔM _VEi determined and the input signal x _i in the shift register 56 saved as described above.

At 74 the change is made by the test torque pulse ΔM _VEsi in the actual torque curve Mv best compensated by the electric machine 4th and the EM counter torque signal ΔM _EMi certainly. Then the dynamic correction value a _dki subtracted and the output signal y _j formed and in the shift register 58 saved as described above.

At 75 it is checked whether the release conditions are still met. If not, step 71 is repeated.

If the result is yes, it is checked at 76 whether the test torque signal has ended. If not, step 72 is repeated.

If 76 is yes, the correlation coefficient becomes CCC calculated at 77 as described above.

At 78 it is checked whether the CCC above the threshold CCCs lies. If not, step 71 is repeated.

If 78 results in yes, the weighting factor is set at 79 b _yx and the more accurate second approach torque signals ΔM _VEik determined as described above.

At 80, the more accurate second approach torque _signals .DELTA.M VEik (k) are indexed sequentially in different operating ranges k determined and in a torque changes correction map 60 registered. Step 71 is then repeated.

List of reference symbols

1

Parallel drive train in P1 arrangement

2

Internal combustion engine

3

crankshaft

4th

Electric machine

5

coupling

6th

transmission

7th

Transmission input shaft

8th

bikes

9

Powertrain control unit

10

Control for the internal combustion engine

11

Control unit for the electric machine

12th

Power electronics

13th

Traction battery

14th

second electric machine

20th

Parallel drive train in P2 arrangement

30th

Serial drive train

50

Processor with integrated memory

51

Target calculation component

52

Test pattern generator

53

Manipulated variable component

54

CAN interface

55

CAN bus system

56

Shift register

57

Dynamics correction component

58

Shift register

59

Correlator

60

Torque changes correction map

M _v

actual torque curve

ΔMv

Change of the actual torque curve

M _EM

EM torque

ΔM _EM

EM torque change

M _Tmin

Transmission input torque

M _VEs

requested target torque signal

ΔM _VEsi

Test torque pulse

M _VEi , ΔM _VEi

first approach torque signal

M _EMi , ΔM _EMi

EM counter torque signal

ΔM _VEid

absolute torque deviation

ΔΔM _VEid

Deviation between torque changes

M _VEik , ΔM _VEik

more accurate second approach torque signal

Xi

Input signal

y _j

Output signal

a _dki

Dynamic correction value

τ

Time Delay

CCC

Correlation coefficient

CCCs

Threshold

b _yx

Weighting factor

fmk

Correction factor map

k

index

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• DE 102015115714 A1 [0008]
• DE 102014214541 A1 [0009]
• DE 102016212113 B3 [0011]
• DE 102015111409 B3 [0012]

Claims (15)

Control for an internal combustion engine in a hybrid vehicle for a correction of a torque determination, which is configured to determine a first approximate torque signal from an actual torque profile of the internal combustion engine, which is established on the basis of a target torque signal, and based on an in / on an electric machine measured EM counter torque signal to correct the first approach torque signal to determine a more accurate second approach torque signal. Control according to Claim 1 , wherein the controller is set up to apply a test torque signal to a requested target torque signal in order to generate the target torque signal. Control according to Claim 2 , wherein the test torque signal consists of several test torque pulses. Control according to Claim 2 or 3 wherein the controller is configured to correct the EM counter-torque signal based on the requested target torque signal in order to obtain a corrected EM counter-torque signal. Control according to claim Claim 4 , wherein the controller further comprises at least one shift register, and the controller is configured to have multiple values of an input signal, which is the sum of the requested target torque signal and the first approximation torque signal, and multiple values of an output signal, which corresponds to the corrected EM counter-torque signal, at points in time to be stored in the shift register within the duration of the test torque signal. Control according to Claim 5 , wherein the times of the values of the input and output signals differ by a predetermined time delay. Control according to Claim 6 , wherein the controller is set up to determine a correlation coefficient from the stored values of the input and output signals. Control according to Claim 7 , wherein the time delay is chosen according to a maximum value of the correlation coefficient. Control according to one of the Claims 6 to 8th , the controller being set up to determine a weighting factor from the stored values of the input and output signals. Control according to Claim 9 , wherein the controller is configured to determine the second approach torque signal based on the stored values of the output signal and the weighting factor. Control according to Claim 10 , wherein the controller is set up to determine the second approach torque signal only when the correlation coefficient is above a predetermined threshold value. Control according to one of the preceding claims, wherein the control is set up to determine the second approach torque signal only when one or more release conditions are met. Drive train for a hybrid vehicle which has a control for an internal combustion engine according to one of the preceding claims. Hybrid vehicle, the hybrid vehicle following a powertrain Claim 13 having. Method for correcting a torque determination in a control for an internal combustion engine according to one of the Claims 1 to 12th , wherein the method comprises: determining a first approximate torque signal from an actual torque profile of the internal combustion engine, which occurs on the basis of a setpoint torque signal; and correcting the first approach torque signal based on an EM counter torque signal measured in / on an electric machine to determine a more accurate second approach torque signal.

Images