Method and system for predicting engine starting performance of an electrical energy storage system

Abstract

The invention relates to a method for predicting the engine starting performance of an electrical energy storage system, in particular a battery for starting a motor vehicle. The method comprises the following method steps: generating engine starting data as a characteristic of the electrical energy storage system; evaluating the generated engine starting data; and outputting an evaluation result, which relates to a prediction of the engine starting performance of the electrical energy storage system. According to the invention, it is in particular provided that the vehicle manufacturer, vehicle model and / or vehicle variant of the vehicle to be started by the electrical energy storage system is taken into account for the evaluation of the generated engine starting data.

Description

[0001] The present invention generally relates to methods and systems for estimating the state of an electrical energy storage system, such as estimating the remaining capacity of an electrical energy storage system. In particular, the present invention relates to methods and systems for predicting the engine starting performance of an electrical energy storage system, especially a motor vehicle starting battery.

[0002] The term "engine starting performance" as used herein is understood to mean the ability of the electric energy storage system to start, in particular to cold start or hot start, a vehicle engine of a motor vehicle. In particular, the term "engine starting performance" also includes the ability of the electric energy storage system to still successfully perform a certain number of starting procedures of the vehicle engine.

[0003] As more and more vehicle functions require at least electrical energy to support them, determining the state of the electrical energy storage system and in particular the state of the vehicle is becoming increasingly important. As more and more safety-relevant functions in the vehicle are supported electrically, the reliability of the state determination and state assessment of the energy storage system is becoming increasingly important.

[0004] In an energy storage system with a plurality of rechargeable battery cells, the individual battery cells age over time, whereby the characteristics of the battery cells or the entire energy storage system change. In order to be able to reliably predict the remaining service life (battery life) of the energy storage system during operation and to be able to identify aging processes in a timely manner and to be able to take appropriate measures, it is desirable to be able to determine the health state of the energy storage system as accurately as possible. Appropriate measures include, for example, replacing aged battery cells or the entire energy storage system before their state becomes critical.

[0005] Known systems and methods for determining battery status generally use direct measurements of the battery status and / or battery characteristics to determine the battery status.

[0006] For example, publication DE 19 540 827 C2 specifies a method for determining a battery state, in which a battery-specific battery aging characteristic diagram is used to determine a battery aging value from detected instantaneous values ​​of variables that are influenced by battery aging.

[0007] For example, document DE 10 335 928 A1 also discloses a method for determining a characteristic variable related to the state of charge of a battery, wherein a first state of charge value based on the high voltage of the battery and a second state of charge value based on the converted charge quantity are determined. The two state of charge values ​​are evaluated with respect to their change in the state of charge in order to derive a characteristic variable of the battery state from the change in the state of charge.

[0008] A disadvantage of these known systems is that large errors in the measured variables directly influence the output variables and therefore the determination of the state of the energy storage system is not unambiguous or may be distorted.

[0009] In addition, battery testing devices are used, which can be used to predict the amount of charge that a rechargeable battery can store. The capacity that can be extracted from a battery depends on the discharge process, i.e. on the discharge current, the end-of-discharge voltage (= voltage at the end of the discharge) and the extent of the discharge. Depending on the discharge process, rechargeable batteries have different capacities. Therefore, in important indications of the nominal capacity, the discharge current and the end-of-discharge voltage must be specified.

[0010] In general, the capacity that can be drawn from a battery decreases as the discharge current increases. This effect is described by the Pouchert equation. One reason for this is that the increased voltage drop across the battery's internal resistance with increasing current causes a corresponding drop in the output voltage, with the result that the end-of-discharge voltage is reached earlier. In addition to the internal resistance, the finite speed of the electrochemical processes and charge transport processes in the battery are also responsible for the decrease in battery capacity when the charging current increases.

[0011] Conventional battery testing equipment is based on measuring the internal resistance of the battery. Based on this measurement, it is possible to calculate the voltage value that the battery will present when a defined discharge profile or a defined type of discharge is applied. The types of discharge considered are for example discharge with constant current, discharge through constant resistance or discharge with constant power.

[0012] As an alternative to this, the measured battery internal resistance may be used to calculate the cold cranking current and define a threshold at which the battery should be replaced. As an example, the threshold may be 50% of the nominal cold cranking current.

[0013] The cold cranking current (often also called the cold test current) is understood as the maximum current that a battery can deliver within 30 seconds at a temperature of -18°C before the battery voltage drops too low. According to the German Industrial Standard (DIN), a fully charged 12V battery discharged to 6V at -18°C should still have a voltage of at least 9V after 30 seconds, and the limit value of 6V is reached only after 150 seconds. According to the American standard (SAE), a fully charged battery should still have a voltage of at least 7.2V after a discharge of 30 seconds at -18°C. According to the IEC (International Electrotechnical Commission), a battery should still have a voltage of 8.4V after a discharge of 60 seconds at -18°C. Last but not least, according to the European standard (EN), the discharge time after discharge to a final voltage of 7.5V should be at least 10 seconds.

[0014] However, conventional methods for estimating or predicting the capacity that can be drawn from a battery are based on the assumption that the battery is operated at a fully charged state. Boundary conditions are often no longer achievable, especially in the case of start-stop batteries, since these batteries are often operated at a partial state of charge (typically 70% to 80%).

[0015] In addition, conventional battery testers do not take battery degradation into account. The term "degradation" is to be understood as a decrease in the capacity of a battery over time, even with correct use, this decrease in capacity occurring due to chemical reactions (aging). On the one hand, the charging and discharging processes at the battery electrodes lead to (only partially reversible) electrochemical processes which prevent complete charging or discharging. For lead-acid rechargeable batteries, these electrochemical processes include sulfation, for batteries based on nickel technology, for example, battery inertia effects, and for batteries based on lithium chemistry, electrode aging due to irreversible parasitic chemical reactions (calendar life).

[0016] The corresponding wear of a rechargeable battery during use is related to the charge capacity and therefore also to the energy density.

[0017] Due to the growing trend towards the use of so-called start-stop systems or automatic start-stop systems, the disadvantages of the known systems for estimating the battery capacity discussed above are no longer acceptable. This is particularly because in a start-stop system, in order to reduce fuel consumption when stationary (e.g. when stopped at traffic lights), the engine is shut down and automatically restarted when stopped under certain conditions. These start-stop systems therefore require the start-stop battery to be in a specific state.

[0018] A start-stop battery performs almost the same tasks and functions as a normal car battery, but it has to do more because the automatic start-stop system causes the internal combustion engine to constantly turn on and off, which requires the battery to perform at its maximum performance. Therefore, it is obvious that the automatic start-stop system requires a battery with sufficient power to work properly.

[0019] The object of the present invention is to specify an optimized diagnostic and analysis cell in order to be able to quickly and reliably predict possible battery defects, in particular of start-stop batteries.

[0020] In particular, a method for estimating the state of health of a battery system, in particular a method for predicting the engine starting performance of a battery system, in particular a battery for starting a motor vehicle, is provided. Using this method, the state of health of the battery system can be estimated as accurately as possible, and in particular the engine starting performance of the battery system can be predicted as accurately as possible.

[0021] A further object of the present invention is to provide a method which can be used for a large number of different energy storage systems, in particular battery systems, and which also provides reliable results under changing operating conditions of the energy storage system or battery system.

[0022] In order to solve this problem, a method according to independent patent claim 1 and a system according to independent patent claim 16 are provided according to the invention.

[0023] Accordingly, the subject matter of the invention is in particular a method for predicting the engine starting performance of an electrical energy storage system, in particular a battery for starting a motor vehicle, with which method not only the specific state of health (SOH) of the energy storage system can be estimated, but also the engine starting performance of the energy storage system can be reliably predicted. For this purpose, it is provided according to the invention that engine starting data are first generated as a characteristic of the electrical energy storage system. These generated engine starting data are evaluated accordingly and an evaluation result is then output, which relates to a prediction of the engine starting performance of the electrical energy storage system. In particular, in order to evaluate the generated engine starting data, the vehicle manufacturer, vehicle model and / or vehicle variant of the vehicle to be started by the electrical energy storage system are taken into account.

[0024] The basis for the method according to the invention thus forms a practical field of application for checking an electrical energy storage system for engine starting performance. This makes use of the knowledge that, depending on the respective field of use or application, the energy storage system still provides sufficient engine starting performance to be able to successfully start a specific vehicle despite a poor health state. The term "specific vehicle" as used herein is to be understood in particular as a specific vehicle manufacturer, a specific vehicle model and / or a specific vehicle variant.

[0025] In other words, according to the invention, engine start data are generated in a large number of different application situations, the engine start data representing the characteristics of the electrical energy storage system checked for engine start performance. For example, the characteristic engine start data are the voltage behavior in the energy storage system during the engine start phase occurring in a specific vehicle (i.e., in a specific vehicle manufacturer, a specific vehicle model and / or a specific vehicle variant).

[0026] In particular, the engine start data as a characteristic of the electrical energy storage system comprises at least one engine start voltage and / or at least one engine start voltage profile of the electrical energy storage system during a start phase of the specific vehicle.

[0027] In a variation thereof, the engine starting data as a characteristic of the electrical energy storage system comprises a temperature of the electrical energy storage system when at least one engine starting voltage is generated and / or when at least one engine starting voltage profile is generated.

[0028] Alternatively or additionally, the engine starting data as a characteristic of the electrical energy storage system comprises state of charge data of the electrical energy storage system when at least one engine starting voltage is generated and / or when at least one engine starting voltage profile is generated.

[0029] In principle, it is conceivable that engine start data as characteristics of the electric energy storage system comprise, for example, at least one minimum value of the voltage of the electric energy storage system during engine start, an engine start time that in particular depends on the state of charge of the electric energy storage system, and / or the number of engine starts that have been performed by the electric energy storage system.

[0030] In this case, it is fundamentally proposed that the engine start data as a characteristic of the electric energy storage system are generated by at least one sensor which is arranged specifically in the vehicle to be started by the electric energy storage system. For example, it is thus conceivable that the engine start data as a characteristic of the electric energy storage system are generated and / or provided for evaluation by a (vehicle-internal) vehicle diagnostic system of the vehicle to be started by the electric energy storage system.

[0031] As an alternative or in addition, the engine start data as a property of the electrical energy storage system may be generated and / or provided by at least one sensor which is preferably directly galvanically connected to the electrical connections of the electrical energy storage system.

[0032] According to an improvement of the method according to the invention, provision is made for taking into account not only the vehicle (vehicle model, vehicle manufacturer and / or vehicle variant) to be started by the energy storage system to be checked, but also the temperature or temperature range of the vehicle engine to be started by the energy storage system, wherein the output results of the evaluation related to the prediction of the engine starting performance of the energy storage system are obtained at different temperatures of the vehicle engine.

[0033] For example, it is envisaged that the method according to the invention could be used to predict the engine starting performance of an electrical energy storage system in summer or winter.

[0034] The term "engine starting performance" of an electric energy storage system as used herein is generally understood to be an indicator of the ability of the electric energy storage system to start, in particular cold start or hot start, a vehicle engine, taking into account the vehicle manufacturer in the evaluation, taking into account the vehicle model in the evaluation and / or taking into account the vehicle variant in the evaluation. For example, the engine starting performance of the electric energy storage system indicates the number of vehicle engine starting procedures that can still be successfully performed by the electric energy storage system.

[0035] According to the invention, a prediction model or aging model for the electrical energy storage system is created and / or continuously optimized based on the training data. For this purpose, the engine start data are input into an engine start prediction algorithm based on machine learning principles to evaluate the generated engine start data. In the engine start prediction algorithm, the engine start data can be divided into different categories according to a classification, in which various modes are distinguished. In particular, in this case, the engine start data are divided into different categories depending on the vehicle manufacturer, vehicle model and / or vehicle variant of the vehicle to be started by the electrical energy storage system.

[0036] In the method according to the invention, provision is made in particular for a learning phase of the engine start prediction algorithm to be carried out. In the learning phase, learning data are input into the engine start prediction algorithm, wherein the engine start prediction algorithm recognizes patterns and / or regularities in the input learning data according to machine learning principles and uses these patterns and / or regularities accordingly when evaluating the generated engine start data. The learning data in particular include characteristic engine start data of a large number of electrical energy storage systems, in particular those electrical energy storage systems from a large number of different vehicle manufacturers, vehicle models and / or vehicle variants that have aged to varying degrees.

[0037] According to a preferred development, provision is made that during a learning phase of the engine start prediction algorithm, engine start data generated by the electrical energy storage system whose engine start performance is to be predicted are used as learning data.

[0038] The invention also relates to a system for predicting the engine start performance of an electrical energy storage system, wherein the system comprises an input interface for inputting engine start data as a characteristic of the electrical energy storage system and an evaluation device for evaluating the input engine start data. In this case, the evaluation device is designed to take into account the vehicle make, vehicle model and / or vehicle variant of the vehicle started by the electrical energy storage system when evaluating the engine start data. In particular, the evaluation device comprises an engine start prediction algorithm based on machine learning principles.

[0039] Exemplary embodiments of the method according to the invention or the system according to the invention will be described in more detail below with reference to the accompanying drawings.

[0040] In the attached picture:

[0041] Figure 1 A flow chart of an exemplary embodiment of a method according to the invention is schematically shown.

[0042] Traditional battery test systems attempt to predict the ability of an energy storage system or battery to start an internal combustion engine. This is usually based on a measurement of the internal resistance (IR) of the energy storage system or a measurement of the electrical conductivity. Based on this measurement, it is possible to calculate the voltage value that the energy storage system will have when a defined current profile is applied.

[0043] The second approach is to use the internal resistance of the energy storage system to calculate the so-called cold cranking current (CCA) and set a threshold value (e.g. 50% of the nominal CCA value) at which the energy storage system should be replaced. The disadvantages of conventional battery testing equipment can be seen in particular, since this type of prediction only takes into account the performance of the energy storage system under laboratory conditions at predefined current values ​​and fixed (estimated) threshold values ​​(such as, for example, 50% of the nominal CCA value). Therefore, conventional battery testing equipment cannot reflect "real" engine starting conditions.

[0044] Taking into account all the variations in vehicle manufacturers and vehicle models, variations in battery temperature and ambient temperature, taking into account the impact of SOC variations, etc., traditional battery testing methods or battery testing equipment can actually only provide an inadequate prediction of battery performance for the actual engine starting process in a vehicle.

[0045] In particular, the methods known from the prior art for predicting battery performance do not take into account the main cause of battery failure, namely the age-related decrease in the battery performance for starting the internal combustion engine.

[0046] In contrast, according to the present invention, vehicle-based electric power consumption is used during engine starting to adjust predictions of the engine starting performance of the energy storage system.

[0047] In particular, the present invention takes into account the differences between current battery testers developed to estimate battery performance under laboratory conditions and the actual battery power required to start a vehicle's internal combustion engine (for different vehicle manufacturers, vehicle models, engine technologies (e.g., diesel, gasoline, start / stop, etc.)).

[0048] like Figure 1 As schematically shown, the method according to the invention is based in principle on two combined methods.

[0049] The first step A of the method forms the basis for the vehicle start prediction algorithm. A large number of different vehicle makes and vehicle models are used for the implementation of method step A (eg 100 vehicles).

[0050] As indicated by box A1 “Vehicle data recording”, different voltage profiles recorded at engine start are used as input data, ie for a large number of batteries at different ages, different temperatures and different SOC conditions.

[0051] In the second step (Box A2 - "Data Analysis"), an intelligent machine learning algorithm classifies the different voltage profiles (e.g., minimum voltage, start time, number of successful engine starts, etc.) of different vehicle makes and vehicle models, and defines performance constraints for successful and unsuccessful engine start processes.

[0052] The big advantage of this approach is that the vehicle's engine start reactions to different variations (vehicle make and model, temperature, SOC, etc.) can be used later in the battery tester for battery replacement recommendations, rather than the unknown battery internal (chemical) reactions.

[0053] Battery testing equipment may record engine start data from an internal combustion engine directly at the energy storage system's terminals, or it may record these values ​​at other points (e.g., on a vehicle's OBD port via analog output pins).

[0054] The engine start data is then input into an engine start prediction algorithm based on machine learning principles to evaluate the resulting engine start data (box A3 - "Engine Start Prediction Algorithm").

[0055] In the engine start prediction algorithm, the engine start data is divided into different categories according to a classification, which have different characteristic patterns (box A4 - "Classification"), where the engine start data is then divided into different categories depending on the vehicle manufacturer, vehicle model and / or vehicle variant of the vehicle to be started by the electrical energy storage system (box A5 - "Evaluation").

[0056] Since predictions that are as reliable as possible require basic algorithms based on the fundamental principles of machine learning, it is advantageous to have as much learning data as possible available to the algorithms.

[0057] For this purpose, a further method step (step B) is used which requires a cloud-based data structure which contains all values ​​defined by the first method step A. The idea is that the battery tester contains the basic algorithm in the first step and works with the Beta version during a pilot or learning phase and collects further engine start data in a number of rounds which are automatically sent to the cloud database.

[0058] Specifically, in Figure 1 In the method step B, the cloud-based data structure is indicated by the reference numeral B1. Method step B is characterized in that a learning phase of the engine start prediction algorithm is performed, in which learning data B2 are input into the engine start prediction algorithm, and the engine start prediction algorithm recognizes patterns and / or regularities in the input learning data and applies these patterns and / or regularities accordingly when evaluating the generated engine start data (box B3 - "training").

[0059] The learning data particularly comprises characteristic engine start data of a large number of electrical energy storage systems, particularly those electrical energy storage systems which have aged to varying degrees from a large number of different vehicle makes, vehicle models and / or vehicle variants.

[0060] During the learning phase of the engine start prediction algorithm, the generated engine start data of the electrical energy storage system whose engine start performance is to be predicted may then be used as learning data.

[0061] The machine learning algorithm evaluates these new engine start data and adjusts the display to optimize the engine start prediction algorithm and its thresholds. From time to time (i.e. after a sufficient number of new engine start data for different vehicle models), the "old" algorithm version is updated to the optimized version. The algorithm itself can be used / implemented in the battery tester unit, and / or the algorithm can be used as a cloud-based algorithm.

[0062] In summary, the following points still need to be explained:

[0063] According to the present invention, firstly, engine starting data is generated as a characteristic of the electric energy storage system to be checked.

[0064] These engine start data are preferably generated directly in the vehicle using appropriate sensors (e.g. sensors that are part of the vehicle diagnostic system). Characteristic engine start data are "real" data generated by the energy storage system during a real internal combustion engine start process of different vehicle manufacturers, vehicle models and vehicle variants. This particularly relates to the engine start voltage and / or the engine start voltage profile, but the temperature of the electrical energy storage system when the engine start voltage or the engine start voltage profile is generated, as well as the state of charge of the energy storage system during the engine start phase may also be considered.

[0065] In other words, the input data for the machine learning-based evaluation algorithm are never laboratory data from the energy storage system, but data recorded under real conditions, here in particular taking into account the vehicle make, vehicle model and / or vehicle variant that generated the real engine start data.

[0066] Then, the machine learning-based evaluation algorithm divides the engine startup data, which serves as learning data, into different categories.

[0067] This provides an engine start prediction algorithm which is regularly fed with new learning data to optimize its accuracy based on a learning phase which is preferably performed continuously or periodically.

[0068] Using the evaluation algorithm formed in this way, it is then possible to evaluate the energy storage system particularly accurately with respect to a specific vehicle model or a specific vehicle manufacturer and / or a specific vehicle variant, in particular the question of how well the engine starting performance of the energy storage system is for a specific vehicle manufacturer, a specific vehicle model and / or a specific vehicle variant.

[0069] The invention is not limited to the method schematically shown in the drawings, but results from the summary of all features disclosed herein.

Claims

1. A method for predicting the ability of an electrical energy storage system to start a vehicle engine of a specific vehicle make, specific vehicle model and / or specific vehicle variant, wherein: The electrical energy storage system is a battery for starting a motor vehicle, wherein the method comprises the following method steps: - generating engine start data as a characteristic of the electrical energy storage system; - evaluating the generated engine start data taking into account the vehicle make, the vehicle model, the vehicle variant of the vehicle started by the electrical energy storage system and / or the temperature or temperature range of the vehicle engine started by the electrical energy storage system; and - outputting a result of said evaluation, wherein said outputted result relates to a prediction of the capability of said electrical energy storage system to be able to start a vehicle engine of a specific vehicle make, a specific vehicle model and / or a specific vehicle variant, wherein the output of said evaluation further relates to a prediction of said ability of said electrical energy storage system to start a vehicle engine of a specific vehicle make, specific vehicle model and / or specific vehicle variant at different temperatures of said vehicle engine; and / or Wherein, the ability of the electric energy storage system to start the vehicle engine of a specific vehicle manufacturer, a specific vehicle model and / or a specific vehicle variant refers to the ability of the electric energy storage system to cold start or hot start the vehicle engine of the vehicle manufacturer considered in the assessment, the vehicle model considered in the assessment and / or the vehicle variant considered in the assessment.

2. The method according to claim 1, in, The engine starting data being characteristic of the electrical energy storage system comprises at least one engine starting voltage and / or at least one engine starting voltage profile of the electrical energy storage system.

3. The method according to claim 2, in, The engine starting data being a characteristic of the electrical energy storage system comprises a temperature of the electrical energy storage system when the at least one engine starting voltage is generated and / or when the at least one engine starting voltage profile is generated; and / or Therein, the engine starting data as a characteristic of the electrical energy storage system includes state of charge data of the electrical energy storage system when the at least one engine starting voltage is generated and / or when the at least one engine starting voltage profile is generated.

4. The method according to claim 1, in, The engine start data as a characteristic of the electrical energy storage system comprises at least one minimum value of a voltage of the electrical energy storage system during engine start, an engine start time, and / or a number of engine starts that have been performed by the electrical energy storage system.

5. The method according to claim 4, in, The engine start time depends on the state of charge of the electrical energy storage system.

6. The method according to claim 1, in, The engine start data characteristic of the electrical energy storage system is generated by at least one sensor.

7. The method according to claim 6, in, The at least one sensor is arranged in the vehicle that is started by the electrical energy storage system.

8. The method according to claim 1, in, The engine start data being characteristic of the electrical energy storage system is generated and / or provided for evaluation by a vehicle diagnostic system of the vehicle on which the electrical energy storage system is started, and / or Therein, the engine start data as a characteristic of the electrical energy storage system is generated and / or provided by at least one sensor.

9. The method according to claim 8, in, The at least one sensor is directly galvanically connected to an electrical connection of the electrical energy storage system.

10. The method according to claim 1, in, The ability of the electrical energy storage system to start a vehicle engine of a specific vehicle make, a specific vehicle model and / or a specific vehicle variant is indicated by the number of vehicle engine starting procedures that can still be successfully performed by the electrical energy storage system.

11. The method according to claim 1, in, The engine start data is input into an engine start prediction algorithm based on machine learning principles to evaluate the generated engine start data.

12. The method of claim 1, in, In order to evaluate the generated engine start data, the engine start data is input into an engine start prediction algorithm, in which the engine start data is divided into different categories according to classification, and the categories have different characteristic patterns.

13. The method according to claim 12, in, The engine start data is divided into different categories depending on the vehicle make, vehicle model and / or vehicle variant of the vehicle started by the electrical energy storage system.

14. The method according to claim 12, in, The method further comprises the method steps of performing a learning phase of an engine start prediction algorithm, wherein learning data is input into the engine start prediction algorithm during the learning phase, and wherein the engine start prediction algorithm identifies patterns and / or regularities in the input learning data and uses these patterns and / or regularities accordingly when evaluating the generated engine start data.

15. The method of claim 14, in, The learning data includes characteristic engine starting data of a large amount of electric energy storage system.

16. The method of claim 15, in, The plurality of electrical energy storage systems may be electrical energy storage systems from a plurality of different vehicle manufacturers, vehicle models and / or vehicle variants that have aged to varying degrees.

17. The method according to any one of claims 14 to 16, in, During the learning phase of the engine start prediction algorithm, the generated engine start data of the electrical energy storage system for which engine start performance is to be predicted is used as learning data.

18. A system for predicting the ability of an electrical energy storage system to start a vehicle engine of a particular vehicle make, a particular vehicle model, and / or a particular vehicle variant, wherein: The electrical energy storage system is a motor vehicle starting battery, wherein the system comprises an input interface for inputting engine starting data as a characteristic of the electrical energy storage system and an evaluation device for evaluating the inputted engine starting data, wherein the evaluation device is designed to take into account the vehicle make, vehicle model, vehicle variant of the vehicle started by the electrical energy storage system and / or the temperature or temperature range of the vehicle engine started by the electrical energy storage system when evaluating the engine starting data, and wherein an output of the evaluation relates to a prediction of the ability of the electrical energy storage system to start a vehicle engine of a specific vehicle make, specific vehicle model and / or specific vehicle variant at different temperatures of the vehicle engine; and / or Wherein, the ability of the electric energy storage system to start the vehicle engine of a specific vehicle manufacturer, a specific vehicle model and / or a specific vehicle variant is the ability of the electric energy storage system to cold start or hot start the vehicle engine of the vehicle manufacturer considered in the assessment, the vehicle model considered in the assessment and / or the vehicle variant considered in the assessment.

19. The system of claim 18, in, The evaluation device includes an engine start prediction algorithm based on machine learning principles.

20. The system of claim 18 or 19, in, The system comprises an output interface for outputting a result of the evaluation performed by the evaluation device, and wherein the system is designed to perform the method according to one of claims 1 to 17 .

Images