Method for determining the temperatures of a battery system at a plurality of determination points of the battery system

Abstract

The present invention relates to a method for determining the temperatures of a battery system at a plurality of determination points of the battery system, having the steps of: capturing a plurality of data sets, wherein each of the data sets comprises temperature data at a plurality of measurement points of the battery system and physical parameters of the battery system, wherein the physical parameters comprise a current of the battery system, a voltage of the battery system, a load connected to the battery system, and / or individual temperatures; training the statistical temperature model using the captured data sets by linking the temperature data at the multiple measurement points of the battery system with the physical parameters using machine learning algorithms; and determining the temperatures at the plurality of points of the battery system based on the physical parameters using the statistical temperature model.

Description

[0001] Method for determining temperatures of a battery system at several locations of the battery system

[0002] The present invention relates to a method for determining temperatures of a battery system at a plurality of determination points of the battery system, a method for controlling the temperature of a battery system, a computer program product for carrying out such methods, a device for determining temperatures of a battery system and a device for controlling the temperature of a battery system.

[0003] It is known in the prior art that battery systems comprising multiple individual batteries can be measured using temperature sensors. However, it is difficult to place the temperature sensors everywhere. Analytical models are often used to predict temperature distributions. Furthermore, a thermal circuit is used to cool or heat the battery system. The circuit path within the battery system is selected based on thermal simulation processes. This approach can sometimes lead to efficient cooling or heating, but can also sometimes lead to serious temperature imbalances, as the coolant exchanges heat as it flows and slowly reaches a different temperature than intended.

[0004] Against this background, it is an object of the present invention to at least partially remedy the disadvantages described above.

[0005] In particular, it is an object of the present invention to be able to reliably determine, predict and control the temperature distribution in the battery system.

[0006] The above objects are achieved by a method for determining temperatures of a battery system having the features of claim 1, a method for controlling the temperature of a battery system having the features of claim 6, a computer program product having the features of claim 11, a device for determining temperatures of a battery system having the features of claim 12 and a device for controlling the temperature of a battery system having the features of claim 13. Further features and details of the invention emerge from the subclaims, the description and the drawings.Features and details that are described in connection with the method according to the invention for determining temperatures of a battery system naturally also apply in connection with the method according to the invention for controlling the temperature of a battery system, the computer program product according to the invention, the device according to the invention for determining temperatures of a battery system and the device according to the invention for controlling the temperature of a battery system and vice versa, so that with regard to the disclosure of the individual aspects of the invention, reference is or can always be made to each other.

[0007] Accordingly, a method for determining temperatures of a battery system is described. The method comprises the following steps:

[0008] Creating a statistical temperature model of the battery system,

[0009] Acquiring a plurality of data sets, wherein each of the data sets comprises temperature data at a plurality of measuring points of the battery system and physical parameters of the battery system, wherein the physical parameters comprise a current of the battery system, a voltage of the battery system, a load connected to the battery system, and / or individual temperatures, and

[0010] Training the statistical temperature model using the acquired data sets by linking the temperature data at the multiple measuring points of the battery system with the physical parameters via machine learning algorithms, and

[0011] Determining the temperatures at the multiple locations of the battery system based on the physical parameters using the statistical temperature model.

[0012] The statistical temperature model is used to determine temperatures at multiple measurement points within the battery system. By linking the multiple temperature data from different measurement points within the battery system with the physical parameters, the temperature distribution of the battery system can be determined based on knowledge of the physical parameters, without having to measure the temperature at each of the different measurement points individually. Furthermore, knowledge of the future development of the physical parameters also allows for conclusions to be drawn about the future development of the temperature distribution within the battery system. This allows proactive counteraction against potential temperature changes at specific measurement points.

[0013] This protects individual batteries from a higher degradation rate during fast charging due to temperature differences between them. With temperature differences between individual batteries, the charging current cannot be optimally selected for all individual batteries.

[0014] Temperature data are temperatures with the corresponding information about the respective location of the temperature in the battery system.

[0015] The battery system can consist of a single battery or multiple individual batteries. The battery system can be used particularly in vehicles.

[0016] The statistical temperature model is a machine learning model. In particular, it can be an artificial neural network (ANN). The statistical temperature model is created using artificial intelligence, i.e., machine learning algorithms. To do this, machine learning algorithms build a statistical temperature model based on the data sets as training data. When training the statistical temperature model, the weights of the artificial neural network are adjusted, for example.

[0017] The temperature data can be measured at several measuring points in the battery system so that the best possible spatial temperature profile can be created.

[0018] A data set is a collection of several related physical properties of the battery system, as they are measured at the same time. The multiple data sets are created at different times and under different conditions of the battery system. Ideally, as many data sets as possible are created during real operation. Different application cases of the battery system can be considered. According to one embodiment of the method, the physical parameters include a current of the battery system, a voltage of the battery system, a load connected to the battery system, and / or individual temperatures. The individual temperatures can be measured at different locations within the battery system and / or in the environment of the battery system. This allows individual temperatures to be used to access the temperature distribution of the battery system, i.e.to draw conclusions about the temperatures at the various locations of the battery system using the statistical temperature model.

[0019] According to a further embodiment of the method, the data sets are acquired from multiple battery systems. In particular, the battery systems are identical in design. Accordingly, the statistical temperature model is created for a specific type of battery system.

[0020] The number of multiple battery systems can be between 10 and 1,000, between 10 and 10,000, or between 10 and 100,000. Even higher numbers are also possible.

[0021] According to a further embodiment of the method, the multiple battery systems are part of a test system. The test system is located, in particular, in a test laboratory. The data sets are therefore acquired under laboratory conditions. In particular, the data sets are acquired under conditions that are as realistic as possible, such as simulated operation within a vehicle. The battery systems can undergo various cycles that represent realistic situations, such as the operation of a battery system in a vehicle.

[0022] According to a further embodiment of the method, the multiple battery systems are arranged in multiple vehicles. In particular, one battery system is arranged in one vehicle. The data sets are accordingly recorded from the battery systems of this vehicle fleet. Advantageously, the data sets are then collected under real conditions. This increases the quality of the data sets. According to a further embodiment of the method, the data sets are stored on an external computer. The training of the statistical temperature model is carried out by this external computer. The external computer can be an external server. In particular, it can be an external server that is connected to a computer network, for example a cloud.

[0023] A method for controlling the temperature of a battery system is also described. The statistical temperature model is used to control temperatures at the battery system's multiple destination points. The method comprises the following steps:

[0024] Measuring physical parameters of the battery system,

[0025] Determining the temperatures at the multiple destination points of the battery system by means of the statistical temperature model and by means of the measured physical parameters based on the assignment of the physical parameters of the battery system to the temperatures at the multiple destination points of the battery system in the statistical temperature model, and

[0026] Controlling the temperatures at the multiple destinations of the battery system by cooling or heating the battery system.

[0027] By determining the temperatures at the battery system's multiple destination points using the statistical temperature model, a decision can be made as to whether or not further cooling or heating should be applied at the battery system's multiple destination points. For example, a corresponding threshold can be set to determine when it is too hot or too cold. The temperatures at the battery system's multiple destination points can be controlled accordingly.

[0028] The statistical temperature model is created as described above. As described above, the physical parameters can include a battery system current, a battery system voltage, a load connected to the battery system, and / or individual temperatures. The battery system measurement points are those points in the battery system at which the battery system temperature data are measured.

[0029] The determination points of the battery system are those points of the battery system which are determined by the statistical temperature model.

[0030] According to one embodiment of the method, the statistical temperature model is stored on an external computer. The determination of the temperatures is performed by this external computer. The external computer can be an external server. In particular, it can be an external server connected to a computer network, for example, a cloud.

[0031] Accordingly, the Battery Management System (BMS) is located in the cloud. The physical parameters can be uploaded to the cloud. The temperatures at the battery system's multiple measuring points are determined in the cloud using the statistical temperature model and the measured physical parameters. The results can then be transmitted to a local system, such as a vehicle.

[0032] According to a further embodiment of the method, the statistical temperature model is stored on a local computer. The determination of the temperatures is performed by this local computer. The local computer can be located in a vehicle, for example. Furthermore, the statistical temperature model can be updated via a connection between the local computer and the cloud. The determination of the temperatures at the multiple determination points of the battery system takes place locally using the local computer. This has the advantage that the temperatures can be determined even when a connection to the cloud is currently unavailable.

[0033] According to a further embodiment of the method, the battery system comprises a thermal multi-circuit system with at least one valve. The temperatures at the multiple destination points of the battery system are controlled by means of the at least one valve.

[0034] The thermal multi-circuit system can comprise two or more parallel cooling lines, which are controlled by at least one valve. The valve is controlled according to the temperature determined or predicted by the statistical temperature model and / or a measured temperature.

[0035] Furthermore, several different areas of the battery system can be cooled or heated differently. For this purpose, the battery system can, in particular, have several cooling lines and / or several valves.

[0036] According to a further embodiment of the method, cooling or heating at the plurality of destination points of the battery system is carried out as a precaution if the determination of the temperatures by means of the statistical temperature model predicts that this is advantageous.

[0037] Based on knowledge of the future development of the physical parameters of the battery system, the statistical temperature model can also be used to predict changes in temperatures at the battery system's various destination points. For example, the load connected to the battery system may change. This temperature change can occur even before the predicted

[0038] temperature can be counteracted.

[0039] It is advantageous to cool or heat as a precaution if it is already known that there will be a need for cooling or heating in the future.

[0040] Furthermore, the present invention relates to a computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of a method according to the invention for determining temperatures of a battery system or of a method according to the invention for controlling the temperature of a battery system. Thus, a method according to the invention also provides

[0041] Computer program product has the same advantages as have been explained in detail with reference to an inventive method for determining temperatures of a battery model or an inventive method for controlling the temperature of a battery system.

[0042] Furthermore, a device for determining temperatures of a battery system is described. The statistical temperature model is used to obtain temperatures at multiple locations within the battery system.The device comprises an acquisition module for acquiring a plurality of data sets, wherein each of the data sets comprises temperature data at the plurality of measuring points of the battery system and physical parameters of the battery system, wherein the physical parameters comprise a current of the battery system, a voltage of the battery system, a load connected to the battery system, and / or individual temperatures, and a training module for training the statistical temperature model using the acquired data sets, wherein the training module is configured to link the temperature data at the plurality of destination points of the battery system with the physical parameters via machine learning algorithms in order to obtain the temperatures at the plurality of destination points of the battery system based on the physical parameters using the statistical temperature model.

[0043] The detection module and the training module are designed in particular to carry out a method according to the invention for determining temperatures of a battery system.

[0044] Furthermore, a device for controlling the temperature of a battery system is described. The statistical temperature model is used to control temperatures at multiple locations within the battery system. The device comprises a measuring module for measuring physical parameters of the battery system, a determination module for determining the temperatures at the multiple destination locations of the battery system using the statistical temperature model and the measured physical parameters based on the assignment of the physical parameters of the battery system to the temperatures at the multiple destination locations of the battery system in the statistical temperature model, and a control module for controlling the temperatures at the multiple destination locations of the battery system by cooling or heating the battery system.

[0045] The measuring module, the determination module, and the control module are designed in particular to carry out a method according to the invention for controlling the temperature of a battery system. Further possible implementations of the invention also include combinations of features not explicitly mentioned above or described below with regard to the exemplary embodiments. Those skilled in the art will also add individual aspects as improvements or additions to the respective basic form of the invention.

[0046] Further advantages, features, and details of the invention will become apparent from the following description, in which exemplary embodiments are described in detail with reference to the drawings. They show:

[0047] Fig. 1 is a schematic view of a device according to the invention for determining temperatures of a battery system;

[0048] Fig. 2 is a schematic view of an apparatus according to the invention for controlling the temperature of a battery system;

[0049] Fig. 3 is a schematic representation of a vehicle fleet, a cloud and a battery management system; and

[0050] Fig. 4 is a schematic representation of a thermal multi-circuit system of a battery system.

[0051] Fig. 1 shows a schematic view of a device 10 according to the invention for determining temperatures of a battery system. The device comprises a detection module 20 for detecting multiple data sets and a training module 30 for training the statistical temperature model using the detected data sets. Temperatures at multiple locations of the battery system can be obtained using the statistical temperature model.

[0052] The device 10 is particularly designed to carry out a method according to the invention for determining temperatures of a battery system. The method comprises the following steps.

[0053] In a first step, several data sets are collected. Each data set contains temperature data from the battery system's multiple measuring points and physical parameters of the battery system. The physical parameters include, for example, the battery system's current, the battery system's voltage, a load connected to the battery system, and / or individual temperatures.

[0054] In a second step, the statistical temperature model is trained using the acquired data sets by linking the temperature data at the multiple measuring points of the battery system with the physical parameters via machine learning algorithms in order to obtain the temperatures at the multiple destination points of the battery system based on the physical parameters using the statistical temperature model.

[0055] Accordingly, thanks to the statistical temperature model, all temperatures at the battery system's multiple measurement points no longer need to be measured each time. Knowledge of the physical parameters is sufficient to obtain the temperatures at the battery system's multiple target points.

[0056] Fig. 2 shows a schematic view of a device 100 according to the invention for controlling the temperature of a battery system. Temperatures at multiple determination points of the battery system can be controlled using the statistical temperature model. The device 100 comprises a measuring module 120 for measuring physical parameters of the battery system, a determination module 130 for determining the temperatures at the multiple determination points of the battery system, and a control module 140 for controlling the temperatures at the multiple determination points of the battery system.

[0057] The device 100 is particularly designed to implement a method according to the invention for controlling the temperature of a battery system. The method comprises the following steps.

[0058] In a first step, physical parameters of the battery system are measured. These physical parameters include, for example, the battery system's current, its voltage, a load connected to the battery system, and / or individual temperatures.

[0059] In a second step, the temperatures at the multiple determination points of the battery system are determined using the statistical temperature model and the measured physical parameters, based on the assignment of the physical parameters of the battery system to the temperatures at the multiple determination points of the battery system in the statistical temperature model.

[0060] In a third step, the temperatures at the multiple target locations of the battery system are controlled by cooling or heating the battery system. Cooling or heating in the third step occurs according to the battery system temperatures determined in the second step.

[0061] Fig. 3 shows a schematic representation of a vehicle fleet 42, a cloud 52, and a battery management system 60. The vehicle fleet comprises a plurality of vehicles 40. Each vehicle 40 can, in turn, have a battery system. As provided for in the method according to the invention for determining temperatures of a battery system, the data sets are preferably acquired from a plurality of battery systems. These battery systems can each be arranged in a vehicle 40. Alternatively, the battery systems can be part of a test system. Such a test system can, for example, be arranged in a laboratory.

[0062] The data sets acquired by the battery systems can be stored on an external computer 50. The external computer 50 can be located in the cloud 52. Furthermore, the training of the statistical temperature model can be performed by this external computer 50. Alternatively, a local computer can also be used.

[0063] The inventive method for controlling the temperature of a battery system can also be executed in the cloud 52 on an external computer 50. For this purpose, the statistical temperature model is stored on an external computer 50. Furthermore, the determination of the temperatures at the multiple determination points of the battery system is performed by this external computer 50. Only the cooling or heating itself needs to be performed on the respective battery system.

[0064] Alternatively, the statistical temperature model is stored on a local computer. Accordingly, the temperatures at the battery system's multiple locations are determined using the statistical temperature model and the local computer.

[0065] Furthermore, the battery management system 60 can be located in the cloud 52 or on a local computer. The battery management system 60 can send a control signal 62 to a valve 82. The valve 82 can control a first cooling flow 64 to a first cooling plate 84 and a second cooling flow 66 to a second cooling plate 86. Of course, the battery management system 60 can also control multiple valves 82 and multiple cooling plates 84, 86.

[0066] Fig. 4 shows a schematic representation of a thermal multi-circuit system 80 of a battery system. The thermal multi-circuit system 80 has a heat exchanger 88, a pump 90, a heating or cooling device 92, a valve 82, a first cooling plate 84, and a second cooling plate 86. The valve 82 can be used to control the intensity of the second cooling flow 66 through the second cooling plate 86. The valve 82 also influences the first cooling flow 64 through the first cooling plate 84. In principle, a thermal multi-circuit system 80 can have multiple valves 82 and multiple cooling plates 84, 86.

[0067] The temperatures at the multiple destination points of the battery system can be controlled via one or more valves 82 and one or more cooling plates 84, 86. Cooling or heating of the battery system can be performed as a precautionary measure if the determination of the temperatures using the statistical temperature model predicts that this is advantageous.

[0068] List of reference symbols

[0069] 10 Device for determining temperatures of a battery system

[0070] 20 Recording module

[0071] 30 training modules

[0072] 40 vehicles

[0073] 42 vehicle fleet

[0074] 50 external computers

[0075] 52 Cloud

[0076] 60 Battery Management System

[0077] 64 first cooling flow

[0078] 66 second cooling flow

[0079] 80 thermal multi-circuit system

[0080] 82 Valve

[0081] 84 first cooling plate

[0082] 86 second cooling plate

[0083] 88 heat exchangers

[0084] 90 pump

[0085] 92 Heating or cooling device

[0086] 100 Device for controlling the temperature of a battery system

[0087] 120 measuring module

[0088] 130 Determination module

[0089] 140 control module

Claims

Patent claims 1 . A method for determining temperatures of a battery system at several locations of the battery system, characterized by the steps: Creating a statistical temperature model of the battery system, Acquiring a plurality of data sets, each of the data sets comprising temperature data at a plurality of measuring points of the battery system and physical parameters of the battery system, the physical parameters comprising a current of the battery system, a voltage of the battery system, a load connected to the battery system, and / or individual temperatures, and Training the statistical temperature model using the acquired data sets by linking the temperature data at the multiple measuring points of the battery system with the physical parameters via machine learning algorithms, and Determining the temperatures at the multiple locations of the battery system based on the physical parameters using the statistical temperature model.

2. Method according to claim 1, characterized in that the data sets are recorded from several battery systems.

3. The method according to claim 2, characterized in that the plurality of battery systems are part of a test system.

4. The method according to claim 2, characterized in that the plurality of battery systems are arranged in a plurality of vehicles (40).

5. Method according to one of the preceding claims, characterized in that the data sets are stored on an external computer (50) and the training of the statistical temperature model is carried out by this external computer (50).

6. A method for controlling the temperature of a battery system according to one of claims 1 to 5, at the plurality of destination points, characterized by the steps: Measuring physical parameters of the battery system, Determining the temperatures at the multiple destination points of the battery system by means of the statistical temperature model and by means of the measured physical parameters based on the assignment of the physical parameters of the battery system to the temperatures at the multiple destination points of the battery system in the statistical temperature model, and Controlling the temperatures at the multiple destinations of the battery system by cooling or heating the battery system.

7. The method according to claim 6, characterized in that the statistical temperature model is stored on an external computer (50) and the determination of the temperatures is carried out by this external computer (50).

8. The method according to claim 6, characterized in that the statistical temperature model is stored on a local computer and the determination of the temperatures is carried out by this local computer.

9. Method according to one of claims 6 to 8, characterized in that the battery system has a thermal multi-circuit system (80) with at least one valve (82) and the temperatures at the plurality of destination points of the battery system are controlled by means of the at least one valve (82).

10. The method according to any one of claims 6 to 9, characterized in that cooling or heating at the plurality of determination points of the battery system is carried out as a precaution if the determination of the temperatures by means of the statistical temperature model predicts that this is advantageous.

11. A computer program product comprising instructions which, when executed by a computer, cause the computer to carry out the steps of a method having the features of any one of claims 1 to 5 or of a method having the features of any one of claims 6 to 10.

12. Device (10) for determining temperatures of a battery system at several locations of the battery system, according to one of the Claims 1 to 10, characterized by a detection module (20) for detecting a plurality of data sets, wherein each of the data sets comprises temperature data at the plurality of measuring points of the battery system and physical parameters of the battery system, wherein the physical parameters comprise a current of the battery system, a voltage of the battery system, a load connected to the battery system, and / or individual temperatures, a training module (30) for training the statistical temperature model using the detected data sets, wherein the training module is configured to link the temperature data at the plurality of measuring points of the battery system with the physical parameters via machine learning algorithms, and Determining the temperatures at the multiple locations of the battery system based on the physical parameters using the statistical temperature model.

13. Device (100) for controlling the temperature of a battery system according to one of claims 1 to 10 at the plurality of determination points, characterized by a measuring module (120) for measuring physical parameters of the battery system, a determination module (130) for determining the temperatures at the plurality of determination points of the battery system by means of the statistical temperature model and by means of the measured physical parameters due to the assignment of the physical parameters of the battery system to the temperatures at the plurality of destination locations of the battery system in the statistical temperature model, and a control module (140) for controlling the temperatures at the plurality of destination locations of the battery system by cooling or heating the battery system.

Images