CN111376739A

CN111376739A - Method for operating a vehicle

Info

Publication number: CN111376739A
Application number: CN201911366632.9A
Authority: CN
Inventors: J.吉拉尔德; K.豪尔; B.阿尔特
Original assignee: Robert Bosch GmbH
Current assignee: Robert Bosch GmbH
Priority date: 2018-12-27
Filing date: 2019-12-26
Publication date: 2020-07-07
Also published as: KR20200083280A; JP2020108330A; DE102018251735A1

Abstract

The invention relates to a method for operating a vehicle having a drive train with a plurality of drive units, at least one of which is designed as an electric machine, and an energy storage system for drive units designed as electric machines, wherein a physical model (Mod) is used_Phys) Based on the torque (M) requested for the drive train_dem) Determining an operating limit (M') for the drive unit to be used_min,EM、M´_max,EM) The drive train is mapped in the physical model, wherein the determined operating limits (M') for the drive unit to be used are taken into account by means of an artificial intelligence-based approach_min,EM、M´_max,EM) And determining the state of charge and/or the change in state of charge of the energy storage system for the drive train (△ SOS)And wherein the drive train operates according to the operating strategy (B).

Description

Method for operating a vehicle

Technical Field

The invention relates to a method for operating a vehicle having a drive unit designed as an electric machine, to a computing unit, to a vehicle, and to a computer program for carrying out the method.

Background

In addition to motor vehicles with only one internal combustion engine, there are also more and more motor vehicles with one or more electric drive units in addition to the internal combustion engine. Such vehicles are referred to as so-called hybrid vehicles. Vehicles with only an electric drive are also available.

In vehicles with a plurality of drive units, it is desirable to find an operating strategy that is as optimal as possible in order to distribute the required torque or the required power between the drive units. This should typically be done with the goal of optimizing energy efficiency, for which reason energy storage systems are usually also taken into account. The latter can be important even in purely electrically operated vehicles with only one electric drive unit, but with, for example, a plurality of energy storage units.

Disclosure of Invention

According to the invention, a method for operating a vehicle, a computing unit, a vehicle and a computer program for carrying out the method are proposed with the features of the independent claims. Advantageous embodiments are the subject matter of the dependent claims and the following description.

The method according to the invention is used for operating a vehicle having a drive train with a plurality of drive units, at least one of which is designed as an electric machine, and an energy storage system for the drive units designed as electric machines, or a vehicle having a drive train with a drive unit designed as an electric machine and an energy storage system with two energy storage units for the electric machines.

For example, a purely electric vehicle with only one electric drive unit or electric machine is considered as a vehicle, but a purely electric vehicle with a plurality of, that is to say at least two electric drive units or electric machines, can also be considered as a vehicle. However, hybrid vehicles with an internal combustion engine and one or more electric machines as drive units are also conceivable. For further details or limitations of certain variants on specific vehicles, reference may also be made to the following embodiments. The specific topology of the hybrid vehicle is also not critical at present, that is to say that one or more electric machines can be arranged upstream and/or downstream of the transmission and/or also on the axle and/or on one or more wheels.

The drive unit is understood here to mean, in particular, a drive unit with which a torque for driving at least one wheel of the vehicle can be transmitted to this wheel. In this connection, a motor-operated electric machine is considered in the case of an electric machine. However, it is also expedient for the electric machine to be operated as a generator.

The energy storage system has, in particular, one or more energy storage units, such as batteries, which are typically used in vehicles, in order to buffer electrical energy and to recapture it. Such energy storage systems, in particular in purely electric vehicles, may also have a fuel cell or an internal combustion engine (so-called range extender) as an energy storage unit, which is provided only for generating electrical energy by means of a generator, and by means of which the electric machine and/or a further battery can be supplied with electrical energy.

In hybrid vehicles (in which the state of charge of the energy storage system, in particular of the battery, can be varied) there are several degrees of freedom which can be taken into account when the required torque is to be allocated to the drive unit. In particular, the emissions, in particular carbon dioxide emissions, can be reduced as a result of the degree of freedom of the state of charge or of its change. The internal combustion engine can be operated, for example, at the best possible load point with regard to low emissions. In order to achieve a balance between the torques supplied and requested by the internal combustion engine, the electric machine can be operated as a generator or as a motor, depending on the situation, i.e. the battery is charged or discharged.

In order to always design the operation of the vehicle as efficiently as possible, a physical model can be used in which the drive train and the energy storage system that may be present are mapped. The most efficient operating strategy for the drive train with the respective drive unit as a component can then be determined by means of, for example, a cost function. In such a cost function, the power provided by a battery having a specific state of charge or states of charge and/or specific changes in the state of charge can be correlated with the power provided by the internal combustion engine, wherein the weighting can be carried out appropriately.

This physical model is particularly accurate, which also allows analytical calculations to be made for the optimized operating strategy. However, high computation power and often long computation durations are required for this. Numerical approximation loses accuracy. This may be disadvantageous for real-time calculations, especially as the number of degrees of freedom increases, the model and thus the calculation becomes more and more complex. And often cannot be matched to changing system characteristics.

It is also conceivable to use different types of maps in which different scenarios and situations are stored, so that the best possible operating strategy can be selected therefrom. However, with increasing complexity, particularly high memory requirements are required on the running computing units. There is also usually no way to match the changing system characteristics.

Another possibility is to use artificial intelligence based approaches such as neural networks or so-called gaussian process models. These artificial intelligence based approaches or models typically do not require complex, numerical solutions, but can learn, that is, can also adapt or be adapted to changing system characteristics. However, in these approaches, it is often the case that the hard or extreme limits of the model and the critical system requirements are not or are not met correctly. These limits refer, for example, to technically limited limits for the torque of the electric machine. It is sometimes equally difficult to distinguish between hard limits and only desirable (but not necessarily technically necessary) requirements.

Furthermore, these approaches have the disadvantage of less robustness to new conditions. If, for example, driving situations occur which affect the operating strategy but cannot be detected or mapped by artificial intelligence or underlying data, then unreasonable results may be obtained, some of which cannot even be put into practice physically or technically.

A high degree of complexity may also occur, since the additional degrees of freedom mean that the number of inputs to the model for artificial intelligence also strongly depends on the vehicle topology and the number of degrees of freedom.

According to one aspect of the invention, it is now provided that the operating limits for the drive unit to be used are determined on the basis of the torque required for the drive train (which may be requested or predetermined, for example, by the driver and/or a vehicle auxiliary system) by means of a physical model in which the drive train and, if appropriate, the energy storage system are mapped. The drive unit to be used is here a drive unit which is supposed to participate in the operation. It is not necessary to use all existing drive units. It is also possible to use only one drive unit.

The physical model is mapped here in particular on the coupling between the individual drive units (for example when the two drive units are coupled to one another via a transmission and/or a shaft), the clutch which is present if appropriate, the maximum available and minimum available torques of the individual drive units, and, if present, the design of the transmission in the drive train with the respective gear and its transmission ratio, which is limited. Possible considerations for the energy storage system may be important in the range of limits that may thus influence the available torque, in particular in the electric machine, or result in further limits that may deviate from the technically limited limits of the electric machine itself. In such a physical model, the rotational speed of the drive unit and/or the speed of the vehicle can likewise be taken into account in connection with the gear.

There are operating limits for the drive unit to be used but in particular for the electric machine to be determined in this way, in particular these upper and lower limits of the torque, in particular of the electric machine, which enable the requested torque to be provided taking into account the possible torque of the internal combustion engine (if present).

Furthermore, an operating strategy for the drive train is determined by means of an artificial intelligence-based approach (or such a model) taking into account the determined operating limits for the drive unit to be used and the state of charge of the energy storage system (in particular the state of charge of one or more batteries of the energy storage system) and/or a (possibly desired) change in this state of charge. This is done in particular with or with the aim of optimizing the energy efficiency of the drive train and the energy storage unit, which also covers, in particular, as little emissions as possible of, for example, carbon dioxide and/or other harmful substances. The drive train is then operated according to this operating strategy, which covers, in particular, a corresponding actuation of the drive unit.

The operating strategy here includes, in particular, a specific distribution of the required torque to the drive unit and, if necessary, also to the energy storage unit, which can then be charged, for example.

Examples of KI methods for such approaches using artificial intelligence or for determining such a distribution of torques for so-called P2 hybrid topologies, that is to say topologies with an electric drive unit and an internal combustion engine, are neuronal networks with inputs for the torque required by the drive unit, the state of charge and the rotational speed, and outputs for the optimal torque of the electric machine and the determined costs. The input and output terminals are connected by respective neurons. The gaussian model can for example use the same inputs and outputs as the neural network.

The proposed method thus combines a physical model of the vehicle or of the drive train of the vehicle with an artificial intelligence-based approach. If the physical limits can be predefined particularly easily and quickly by means of the physical model, the best possible or at least one best possible operating strategy can be found on the basis of the limits and within these limits by means of artificial intelligence, wherein no or at least hardly any physically impossible states are obtained here.

The operating strategy for the drive train determined by means of an artificial intelligence-based approach (also referred to as KI model or MI approach) is preferably adjusted by means of a physical model before the drive train is operated according to said operating strategy. This means that the physical model is used again in order to adapt or optimize the operating strategy accordingly, if specific limits should not be followed or if necessary also in order to alleviate critical situations.

The values of the parameters of the operating strategy for the drive train, i.e. for example the share of the torque that a particular drive unit should provide, are advantageously buffered and/or used for learning in an artificial intelligence-based approach. This enables a particularly simple and effective and also continuous improvement of the artificial intelligence-based approach or underlying or used model, in particular when this is done in combination with an adjustment of the operating strategy by means of the physical model.

As already mentioned, a vehicle with a drive train having a transmission with at least two different gears is preferably used. The operating strategy for the drive train is determined by means of an artificial intelligence-based approach, in particular also taking into account the currently used gear. Further degrees of freedom can therefore be taken into account in determining the best possible operating strategy.

The operating strategy for the drive train is preferably also determined for at least one other currently unused gear and taking into account the respective gear by means of an artificial intelligence-based approach, wherein the operating strategy or the extended operating strategy then also covers the predetermination of the gear. The best possible operating strategy can thus be determined for the currently used or selected gear as explained above. Furthermore, the best possible operating strategy can also be determined for one or more other gears which are assumed to be available. If one of the other determined operating strategies is better than the operating strategy for the current gear for a specific reason, the gear can be shifted or a prompt for shifting the gear can be given in a manual transmission.

In particular, it is preferred, as already mentioned, to use a vehicle with a drive train, wherein one of the plurality of drive units is designed as an internal combustion engine. In particular, this relates to a hybrid vehicle, as already mentioned at the outset. In such hybrid vehicles, savings or reduction in emissions can be achieved particularly effectively by using a selection of operating strategies, in particular by using a distribution of torque to the drive units. The internal combustion engine can be operated, for example, at a load point that is as optimal as possible with regard to low emissions. In order to achieve a balance between the torque provided and requested by the internal combustion engine, the electric machine can be operated as a generator or as a motor, depending on the situation, i.e. the battery is charged or discharged.

In this case, it is also expedient to determine the operating strategy for the drive train by means of an artificial intelligence-based approach, taking into account the current rotational speed of the internal combustion engine, since it is particularly simple to adapt the optimum load point.

According to a further aspect of the invention, a method for operating a vehicle having a drive train with a drive unit designed as an electric machine and an energy storage system with two different energy storage units for the electric machine is also proposed. These energy storage units can, as already explained above, for example relate to two batteries or to a battery and a fuel cell or a range extender. In this case, the operating limits for the energy storage unit to be used are determined by means of a physical model (in which the energy storage system is mapped) on the basis of the torque requested for the drive train. By means of an artificial intelligence-based approach, an operating strategy for the energy storage unit is then determined to supply electrical energy, taking into account the determined operating limits for the energy storage unit to be used. The energy storage unit then moves according to the operating strategy.

The operating strategy can be used in the same way as the previously described method, with the only difference that no specific torque is distributed to the plurality of drive units, but the required amount of electrical energy supplied is distributed to the different energy storage units. Of course, the two modes of the method can be combined with one another, if a plurality of drive units and a plurality of energy storage units are provided.

Advantageous embodiments of the first aspect of the invention are also applicable to this aspect, mutatis mutandis.

The computing unit according to the invention, for example a control unit of a motor vehicle, is provided, in particular, in terms of program technology, for carrying out the method according to the invention.

The invention also relates to a vehicle having a drive train with a drive unit or a plurality of drive units, which are designed as electric motors, one of which is designed as an electric motor, and an energy storage system for the drive unit, which is designed as an electric motor, and having a computing unit according to the invention. Preferred embodiments and advantages with regard to the vehicle can be seen here with reference to the previously described embodiments of the method, which are correspondingly suitable here.

The implementation of the method according to the invention in the form of a computer program or a computer program product with computer code for carrying out all method steps is advantageous, since this results in particularly low costs, in particular when the controller being implemented can also be used for other tasks and is therefore already present. Suitable data carriers for supplying the computer program are, in particular, magnetic, optical and electrical memories, such as a hard disk, flash memory, EEPROM, DVD, etc. The program can also be downloaded via a computer network (internet, intranet, etc.).

Advantageous advantages and design aspects of the invention result from the description and the drawing.

The invention is illustrated schematically in the drawings by means of embodiments and is described hereinafter with reference to the drawings.

Drawings

Fig. 1 schematically shows a vehicle in which a method according to the invention can be carried out;

FIG. 2 shows a schematic representation of the procedure of the method according to the invention in a preferred embodiment;

fig. 3 shows a schematic representation of the procedure of the method according to the invention in a further preferred embodiment.

Detailed Description

Fig. 1 schematically shows a vehicle 100 according to the invention in a preferred embodiment, in which the method according to the invention can be carried out.

The vehicle 100 has two

axles

110 and 120, the axle 120 being connected to the drive train 101 as a drivable axle with wheels that can be driven in each case. The vehicle 100 or the drivetrain 101 has a drive unit 130 in the form of an internal combustion engine and a drive unit 140 in the form of an electric motor, which are connected in a torque-transmitting manner by means of a clutch 131.

Furthermore, an energy storage system 150, which is designed as a battery or has a battery, is provided and is electrically connected to the electric machine 140. A transmission 160 is also provided in the drive train 101, by means of which different gears can be set or selected.

Furthermore, a computer unit 180 is provided, which is designed as a controller and by means of which the drive unit, the clutches and, if necessary, the transmission can be actuated. Of course, a plurality of computing units can also be provided for this purpose, which are in communication with one another.

The vehicle 100 thus relates to a hybrid vehicle. The invention should be explained here by way of example. Of course, as mentioned at the outset, other types of vehicles or topologies may also be used.

Fig. 2 schematically shows the process of the method according to the invention in a preferred embodiment. For this purpose, a physical model Mod of a drive train with an internal combustion engine and an electric machine, as shown by way of example in fig. 1, is shown_Phys。

In this physical model Mod_PhysThe minimum technically possible torque M of the drive train, for example by means of an electric machine_min,EMAssociated maximum torque M_max,EMMinimum technically possible torque M of the internal combustion engine_min,ICEAssociated maximum torque M_max,ICEAnd requested torque M_demMapped or shown. The maximum and minimum values are limited, for example, by the manner of construction and can be set as parameters. They play a limiting role in the model in particular. It is conceivable, as already mentioned at the outset, for the energy storage system to also be taken into account in order to be able to change the limit for the torque of the electric machine in particular.

With the aid of this physical model Mod_PhysThe requested torque M can then be addressed_demThe operating limits for the drive unit to be used are determined, and the physical model can be adapted to the other drive trains or topologies of the vehicle particularly simply and quickly. In the example shown, these operating limits are the minimum torque M' of the motor_min,EMAnd an associated maximum torque M ″_max,EMOn the basis of which the requested torque M can also be (also) met in consideration of the torque limit of the internal combustion engine_dem。

For example, a minimum torque M' is obtained_min,EMAs at the requested torque M_demAnd maximum torque M_max,ICEWherein this difference is limited to the smallest possible torque M at the electric machine_min,EMAnd maximum torque M of the motor_max,EMA value in between.

E.g. a positive requested torque M_demObtaining the maximum torque M_max,EMAs at the requested torque M_demAnd minimum torque M_min,ICEWherein this difference is limited to the smallest possible torque M at the electric machine_min,EMAnd maximum torque M of the motor_max,EMA value in between.

Also for example, a negative torque M requested_demObtaining the maximum torque M_max,EMAs the requested torque M_demThe torque requested is limited to the minimum possible torque M at the electric machine_min,EMAnd maximum torque M of the motor_max,EMA value in between.

It is also conceivable here to take account of possible states of charge of the energy storage system and/or possible changes of the same energy storage system, since this state of charge can influence the torque that can be output by the electric machine. Conversely, technically limited limits for the torque are typically applied to internal combustion engines.

These operating limits M' of the state of charge of the energy storage system_min,EMAnd M ″_max,EMAnd (expected and/or actual) change △ SOC, the (current) speed n of the internal combustion engine_ICEAnd requested torque M_demThe model Mod can then also be used here_AIWithin the framework of the artificial intelligence-based approach shown, this is processed or taken into account in order to obtain the best possible operating strategy B, as already explained in detail at the outset.

By means of a physical model, the model Mod can thus be generated_AIOr artificial intelligence based approaches reserve a (restricted) solution space, so that possible solutions that are not physically or technically fulfilled are excluded from the outset or at least become impossible.

Is thus determined or obtainedThe operating strategy B here comprises in particular the requested torque M_demDistribution to electric machines and internal combustion engines. In this case, it is also conceivable that, as a whole, a higher torque than requested, for example, is generated, but that this excess torque is used to charge the energy storage system or the corresponding battery.

In addition, the physical model Mod can be used_PhysThe determined or derived operating strategy B is once again adjusted or corrected in order to take into account, for example, security-critical aspects, so that, for example, technically feasible but undesirable behavior can be avoided. Furthermore, the contained values of the operating strategy B or of the parameters can be stored before and/or after the adjustment in the characteristic field K or in another memory unit in order to use them for the subsequent adaptation of the model Mod_AIAnd (4) training.

The proposed practice and further aspects should be summarized once more next. Part of the method with artificial intelligence, that is to say the model Mod_AIThe fuel efficiency results can be based on different application scenarios of the electric machine and the internal combustion engine, in particular also for different driving cycles. It is possible or desirable to use a model Mod with these training data_AIIn order to determine an optimum torque distribution between the electric machine and the combustion engine.

A combined approach consisting of a physical model and an artificial intelligence based model can be used in order to cover all scenarios. The physical model should be used to determine the maximum and minimum possible torque or corresponding power of the electric machine(s) in the system in order to meet the driver demand.

The results of the artificial intelligence-based model can additionally be fed into the physical model, so that additional criteria, such as a higher level of comfort in the case of a gear change, can be taken into account there, as will be explained later, and critical system requests can be checked. For example, the torque of the electric machine, as obtained from an artificial intelligence-based model, can be modified in such a way that the torque requested by the driver is met within the limits of the electric machine itself and within the limits of the internal combustion engine.

The torque of the motor may also be modified in the physical model based on the state of charge of the battery to ensure that the overall system remains within a stable operating range. Physical models may also store data for situations where the results of artificial intelligence based models violate critical system requirements and these stored data can be used to train the models better in the future. The stored inputs can thus be processed separately in the next training result in order to ensure that the training algorithm explicitly covers such combinations of inputs.

An ideal solution for these key combinations of inputs may be determined for offline training and used for training. For offline training, at least the corrected output may be used in the running computing unit for the training.

Fig. 3 schematically shows a flow of the method according to the invention in a further preferred embodiment. In the upper region, the procedure already shown in fig. 2 and explained in detail there is shown in order to find the optimum operating strategy B. However, this applies only to specific gears G₁If gear selection is possible.

The embodiment shown here can therefore be used for another gear G₂The same is done accordingly. Since this is essentially the same as for gear G already explained with reference to fig. 2₁Is not different, so that this is only shown schematically in rough detail here.

The difference being at most in the case of gear G₂Is not used, but rather an assumed speed n' is used instead of the true or current speed_ICEAs rotational speed of the internal combustion engine, e.g. for the gear G concerned₂With the same general conditions. This can be calculated very simply on the basis of the gear ratios of the gears.

Then for gear G₂An operating strategy B' was obtained. Then by comparison for gear G₁Operating strategy B and for gear G₂The operating strategy B' of (a) can in particular be determined within the scope of the extended operating strategy whether a gear should be shifted if necessary. Also contemplated herein areThe time required for shifting gears and/or the possible consequent comfort disadvantage. Of course, other gears are also considered in this way.

Another preferred aspect of the proposed method is that it makes it possible to train artificial intelligence-based models or corresponding functions in the operating computing unit for each specific vehicle in order to take account of the variability of the individual component powers and of the aging behavior of the vehicle. Power information, such as true fuel consumption compared to conditions for a particular vehicle, may be stored, and a data-based model (e.g., a neural network) may be additionally trained. Such training may be performed either directly within the operating computing unit or offline in the cloud or remote computing system so that the adjusted functionality can be re-fed into the vehicle.

The stored data for all cases in which the output of the artificial intelligence based model must be modified in order to meet the critical system requirements can also be used here in online training.

Claims

1. Method for operating a vehicle (100) having a drive train (101) with a plurality of drive units (130, 140), at least one of which is designed as an electric machine, and an energy storage system (150) for drive units (140) designed as electric machines,

in which the physical model (Mod) is used_Phys) Based on a requested torque (M) for the drive train (101)_dem) Determining an operating limit (M') for the drive unit to be used_min,EM、M´_max,EM) A drive train (101) mapped in the physical model,

wherein the determined operating limits (M') for the drive units to be used are taken into account by means of an artificial intelligence-based approach_min,EM、M´_max,EM) And determining an operating strategy (B) for the drive train (101) in the event of a change in the state of charge (△ SOS) and/or the state of charge of the energy storage system (150), and

wherein the drive train (101) is operated according to an operating strategy (B).

2. The method according to claim 1, wherein the operating strategy (B) for the drive train is determined by means of an artificial intelligence-based approach while optimizing the energy efficiency of the drive train (101) and the energy storage unit (150).

3. Method according to claim 1 or 2, wherein the operating strategy (B) for the drive train determined by means of an artificial intelligence-based approach is determined by means of the physical model (Mod) before the drive train (101) is operated according to the operating strategy (B)_Phys) To be adjusted.

4. Method according to any of the preceding claims, wherein values for parameters of the operating strategy (B) of the drive train are buffered and/or used for learning in an artificial intelligence based approach.

5. Method according to any one of the preceding claims, wherein a vehicle (100) with a drive train (101) having a transmission (160) with at least two different gears is used.

6. Method according to claim 5, wherein the currently used gear (G) is additionally taken into account by means of an artificial intelligence-based approach₁) The operating strategy (B) for the drive train is determined.

7. Method according to claim 6, wherein at least one further currently unused gear (G) is also targeted by means of an artificial intelligence-based approach₂) And determining an operating strategy (B') for the drive train taking into account the respective gear, and wherein the operating strategy comprises a pre-specification of the gear.

8. The method according to any one of the preceding claims, wherein one of the plurality of drive units is configured as an internal combustion engine.

9. Method according to claim 8, wherein the current rotational speed (n) of the internal combustion engine is additionally taken into account by means of an artificial intelligence-based approach_ICE) The operating strategy (B) for the drive train is determined.

10. The method according to any one of the preceding claims, wherein the operating strategy (B) determined by means of an artificial intelligence-based approach comprises a distribution of torque to a plurality of drive units.

11. Method according to one of the preceding claims, wherein, instead of a vehicle (100) with a drive train (101) which has a plurality of drive units (130, 140) and at least one of the drive units is configured as an electric machine, and an energy storage system (150) for the drive unit (140) configured as an electric machine, a vehicle (100) with a drive train (101) which has a drive unit configured as an electric machine and an energy storage system (150) with two energy storage units for the electric machine (140) is operated,

in which the physical model (Mod) is used_Phys) Based on a requested torque (M) for the drive train (101)_dem) Determining operating limits for the energy storage units to be used, the energy storage system being mapped in the physical model,

wherein an operating strategy (B) for the energy storage unit for providing electrical energy is determined by means of an artificial intelligence-based approach taking into account the operating limits determined for the energy storage unit to be used, and

wherein the energy storage unit is operated according to an operating strategy (B).

12. A computing unit (180) configured to perform all method steps of the method according to any one of the preceding claims.

13. Vehicle (100) having a drive train with a drive unit or a plurality of drive units (130, 140) designed as electric machines, at least one of which is designed as an electric machine, and an energy storage system (150) for the drive units designed as electric machines, and a computing unit (180) according to claim 12.

14. Computer program which, when being implemented on a computing unit (180), causes the computing unit (180) to carry out all method steps of the method according to any one of claims 1 to 11.

15. A machine-readable storage medium having stored thereon a computer program according to claim 14.