Determining clutch temperature of vehicle clutch by on-demand activation of neural network
By activating the neural network on demand in the transmission controller and using the neural network only to determine the temperature during the vehicle clutch switching process, the problem of waste of computing resources and real-time monitoring in the prior art is solved, and vehicle safety and computing efficiency are improved.
Patent Information
- Application Number
- CN202380044011.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Priority Date
- 2022-06-02
- Filing Date
- 2023-05-23
- Publication Date
- 2024-12-31
AI Technical Summary
The prior art implements neural networks in transmission controllers to determine vehicle clutch temperatures require a lot of computing power, and traditional mathematical models are complex and time-consuming, making it difficult to monitor clutch temperature fluctuations in real time, affecting vehicle safety.
The neural network is activated on demand, and the neural network is activated only during the vehicle clutch switching process. Using models such as multi-layer perceptron, fully connected layer or convolutional neural network, the clutch temperature is determined by identifying the switching process and inputting operating parameters, and Deactivate neural networks when not needed to save computing resources.
The calculation time for determining the clutch temperature is reduced, the calculation ability is allocated on demand is realized, the calculation efficiency of the transmission controller and the driving safety of the vehicle are improved, and the risk of clutch damage caused by rising temperatures is reduced.
Smart Images
Figure CN119234095A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to a method for determining a clutch temperature of a vehicle clutch by means of an on-demand activation of a neural network. The invention also relates to a method for training a neural network, which is designed to determine a clutch temperature of a vehicle clutch. The invention also relates to a subordinate control device for determining a clutch temperature of a vehicle clutch by means of an on-demand activation of a neural network. Background Art
[0002] The temperature of the vehicle clutch can be calculated using conventional rule-based mathematical models. The calculation can take place in the transmission control unit. Machine learning, such as neural networks, can be used for this purpose. However, implementing a neural network on a transmission control unit requires a lot of computing power. Summary of the invention
[0003] In a first aspect, the invention relates to a method for determining a clutch temperature of a vehicle clutch by means of a neural network.
[0004] The vehicle clutch can be installed in a vehicle driven by an engine, such as a motor vehicle, a motorcycle or a two-wheeled vehicle driven at least partially electrically. The vehicle clutch can be used to transmit the driving force applied by the engine of the vehicle to the driving axle of the vehicle. The vehicle clutch can have at least two different switching states or gears, which can be defined by a predetermined ratio between the driven torque of the engine and the driving torque of the driving axle.
[0005] The clutch temperature can be understood as the temperature of at least one clutch element, such as the temperature of a clutch plate. Alternatively, the clutch temperature can also be understood as the temperature of the entire vehicle clutch, which can be based on, for example, determining an average value of the individual temperatures of the clutch elements. Due to the operation of the vehicle, the temperature of the vehicle clutch changes. For example, the temperature of the vehicle clutch will increase during a switching process or gear shift. If the temperature of the vehicle clutch increases above a critical temperature value, this may cause damage to or failure of the vehicle clutch. Therefore, determining the clutch temperature of the vehicle clutch contributes to the safety of the vehicle.
[0006] A neural network can be understood as a mathematical model that at least partially mimics the structure of neurons in the human brain. A neural network can be created with the aid of a computer. A neural network can have an input node, an output node, and a plurality of intermediate nodes, the intermediate nodes being arranged between the input node and the output node. The input node can be, for example, a data interface, via which, for example, input data can be input into the neural network. The output node can be, for example, a data interface, via which data can be output from the neural network. The input node can be connected to the intermediate node, and the intermediate nodes can also be connected to each other. The intermediate node can be connected to the output node. The input data can be historical data detected at a certain point in time. Alternatively or additionally, the input data can be synthetic data generated by processing the detected or measured data. Similarly, the output data can also be historical data or synthetic data.
[0007] Information can be buffered at least temporarily at the intermediate nodes. It can be provided that at least one arithmetic operation is performed at the intermediate nodes. Input data can be transmitted from the input node to the output node via the intermediate node. During the transmission, the input data can be mathematically processed, for example, converted into output data. The intermediate nodes of the neural network can be arranged in one or more layers or planes. The intermediate nodes can be connected to each other within a layer. In addition, the intermediate nodes of a layer can be connected to the intermediate nodes of other layers. The individual connections of the input nodes, the intermediate nodes and the output nodes can be provided with mathematical weights. Depending on the purpose of the neural network, the individual weights of the connections can be different. The weights can be changed during the training of the neural network. By adjusting the mathematical weights of the connections of the individual nodes during the training, the neural network can learn the relationship between the input data and the output data. During the use of the neural network for the set purpose, the neural network can apply the learned relationship to the input data so as to generate output data according to the predetermined purpose of the neural network.
[0008] For determining the clutch temperature, for example, a multilayer perceptron (MLP) can be used as a neural network. The multilayer perceptron has at least one layer of intermediate nodes and uses at least one nonlinear mathematical function to calculate the output data. Another exemplary neural network for determining the clutch temperature can be a fully connected layer (FCL) network. In such a neural network, all input nodes, intermediate nodes and output nodes are connected to one another. Furthermore, for determining the clutch temperature, a convolutional neural network (CNN) can also be used, in which the intermediate nodes of different layers are connected to one another at least partially in the form of a mathematical convolution function.
[0009] The method has a step of identifying a switching process of a vehicle clutch. The switching process can be carried out manually, for example by the driver of the motor vehicle operating a pedal, or automatically by a switching program of the motor vehicle. The switching process of the vehicle clutch can be identified by a change in at least one operating parameter of the vehicle clutch. For example, when the switching process is carried out, the rotational speed of the clutch element of the vehicle clutch changes. Alternatively or additionally, the pressure applied to the clutch element of the vehicle clutch during the switching process changes. Alternatively or additionally, during the switching process, the current intensity of the current flowing through the clutch element of the vehicle clutch changes. Alternatively or additionally, the switching process can be identified by a change in the target gear of the vehicle clutch.
[0010] Furthermore, the method comprises the step of activating the neural network for determining the clutch temperature. The neural network is activated in dependence on the recognition of a switching process. An activation signal can be generated, for example, by a control unit of the vehicle by the recognition of the switching process. The control signal can be used to activate the neural network. For example, the control signal can be transmitted to an activation unit, which can be designed to activate the neural network in dependence on the receipt of the control signal. Activating the neural network can be understood as starting a calculation performed by the neural network. Alternatively or additionally, activating the neural network can be understood as a data access by the neural network to data required for the calculation.
[0011] Furthermore, the method also comprises the step of inputting at least one value of at least one operating parameter of the vehicle clutch as input data into the neural network. The at least one value of the at least one operating parameter of the vehicle clutch can represent the power delivered to the vehicle clutch. Alternatively or additionally, the at least one value of the at least one operating parameter of the vehicle clutch can represent the clutch temperature of the vehicle clutch. The at least one value of the at least one operating parameter can be transmitted to at least one of the input nodes of the neural network, for example, by means of an input device. Alternatively, the at least one value of the at least one operating parameter can be transmitted to at least one of the input nodes of the neural network by a control device of the vehicle via a data interface. The at least one value of the at least one operating parameter can be used as input data for determining the clutch temperature using the neural network. Alternatively or additionally, the at least one value of the at least one operating parameter can be used as input data for training the neural network.
[0012] Furthermore, the method has a step of determining the clutch temperature by means of a neural network. The neural network determines the clutch temperature based on input data and a relationship between at least one operating parameter and the clutch temperature learned by the neural network. The neural network can process the input data, for example at intermediate nodes, according to the learned relationship in order to determine the clutch temperature. When learning the relationship, the neural network can change the mathematical weights of the connections of the individual nodes in order to determine the clutch temperature.
[0013] The method further comprises the step of outputting the clutch temperature determined by the neural network. The determined clutch temperature can be output from the neural network to a device for electrical signal processing configured for this purpose. The clutch temperature determined and output by the neural network can be used by other components of the vehicle control unit. For example, the determined clutch temperature can be used to generate a control value for a control unit of the vehicle, such as a transmission control unit of an automatic transmission. In addition, the control value can also be displayed on a display unit of the vehicle. Alternatively or additionally, the control value based on the clutch temperature can be processed by an evaluation unit, for example, for monitoring the driving safety of the vehicle.
[0014] The method further comprises the step of deactivating a neural network for determining the clutch temperature. The neural network is deactivated in dependence on the determined clutch temperature. Deactivating the neural network may be understood as ending or interrupting the calculation of the clutch temperature by the neural network. Alternatively or in addition, deactivating the neural network may also be understood as ending or interrupting the data access of the neural network to the input data. Deactivating the neural network may be implemented by a control unit of the vehicle configured for this purpose. Alternatively, deactivating the neural network may be implemented by the neural network itself. To this end, it may be provided that the neural network is trained for identifying deactivation conditions for deactivating the neural network. In order to deactivate the neural network, the determined clutch temperature may be processed in order to determine the deactivation conditions that are dependent on the clutch temperature. Alternatively, the determined clutch temperature may be used directly, that is to say without processing, for deactivating the neural network.
[0015] The proposed method enables the clutch temperature to be determined with the aid of a neural network. This makes it possible to dispense with the use of conventional mathematical models, which are generally very complex and computationally very time-consuming. By using a neural network, the computation time required for determining the clutch temperature can be reduced on the one hand. In addition, the proposed method also enables the neural network to be activated on demand for determining the clutch temperature. The neural network is only activated when a switching process of the vehicle clutch is detected. During the switching process, the vehicle clutch is usually subjected to a higher load than during a driving mode in which no switching occurs. Since the vehicle clutch is subjected to a higher load, large fluctuations in the clutch temperature may occur during the switching process. Therefore, the fluctuations in the clutch temperature are to be monitored in particular. On the other hand, if no switching process occurs, the fluctuations in the clutch temperature generally also occur less frequently. Therefore, it is not mandatory to determine or monitor the clutch temperature during this time period. Therefore, the neural network can be deactivated during this time period. The proposed method can be executed, for example, on a transmission control unit. In this case, the on-demand activation of the neural network enables the computing power available on the transmission control unit to be allocated on demand, such as storage bits for calculations and / or computing time for calculations. The neural network calls up these computing powers only when a switching process is detected. If no switching process is detected, the computing power of the transmission control unit can be used for other purposes, for example to ensure the functionality of the transmission.
[0016] According to one embodiment, if the determined clutch temperature is below a defined threshold, the neural network is deactivated. The defined threshold may be defined by a user, such as a driver of the vehicle. Alternatively, the threshold may be defined by the manufacturer of the vehicle. The defined clutch temperature threshold may represent a clutch temperature below which the probability of damage to the vehicle clutch due to temperature rise is sufficiently low. The defined threshold may be based on an empirical value. Alternatively or additionally, the defined threshold may be based on mathematical calculations or experimental tests. The defined threshold may be selected differently for different vehicle clutches. It may be provided that, below the defined threshold, it is no longer necessary to determine or monitor the clutch temperature. For example, below the defined threshold, the possibility of clutch damage due to temperature rise may be fully excluded. If the clutch temperature is below the defined threshold, it is no longer mandatory to determine the clutch temperature by means of a neural network. Therefore, the neural network may be deactivated. Therefore, by determining a defined threshold for the clutch temperature below which the neural network is deactivated, the driving safety of the vehicle may be improved.
[0017] According to another embodiment, the clutch temperature is determined by the neural network within a predetermined time interval. The predetermined time interval can be predetermined by a user, such as the driver of the vehicle. Alternatively, the time interval can be predetermined by the manufacturer of the vehicle. If the clutch temperature is determined by the neural network on the transmission control unit, the time interval can be predetermined depending on the computing power available on the transmission control unit. For example, it can be provided that the neural network uses at most 20% or at most 10% or at most 5% or at most 1% of the total computing time available on the transmission control unit to determine the clutch temperature. As long as the neural network is active, the clutch temperature can be determined by the neural network within the predetermined time interval. As an alternative, the clutch temperature is determined by the neural network within the predetermined time interval a predetermined maximum number of times, for example once or twice or three times. If the predetermined maximum number of calculation processes is reached, it can be provided that the clutch temperature is not further determined by the neural network. By predetermining the time interval within which the neural network determines the clutch temperature, the determination method can be adapted to different types of vehicle clutches or different types of controllers in which the determination is performed. Furthermore, by predefining the time interval, the computing power available for determining the clutch temperature by means of the neural network can be adapted to specific predefined parameters, for example predefined parameters given by a user or a manufacturer.
[0018] According to a further embodiment, the predetermined time interval is predetermined in dependence on the determined clutch temperature. For example, it can be provided that the higher the determined clutch temperature, the shorter the selected time interval. In this case, the clutch temperature can be determined more frequently when the clutch temperature is relatively high than when the clutch temperature is relatively low. The invention does not exclude other forms of determining the time interval in dependence on the temperature. Determining the time interval in dependence on the temperature enables the determination method to be adapted to different operating states of the vehicle clutch.
[0019] According to one embodiment, the method comprises the step of determining at least one input value, which represents the power supplied to the vehicle clutch. The at least one input value is determined based on the processing of temporally consecutive values of the power supplied to the vehicle clutch. The power supplied to the vehicle clutch can be understood as physical power, i.e. energy supplied to the vehicle clutch within a predetermined time period. For example, the power supplied to the vehicle clutch can be the switching power supplied to the vehicle clutch during a switching process or gear shift. The power supplied to the vehicle clutch can represent a change in the clutch temperature of the vehicle clutch.
[0020] The power supplied to the vehicle clutch can be detected at certain time intervals. The individual values of the power supplied to the vehicle clutch can be arranged, for example, in ascending order according to their detection time points. These temporally successive values of the power supplied to the vehicle clutch can be mathematically processed, for example by means of a predetermined arithmetic operation (such as averaging).
[0021] In this embodiment, the method further comprises the step of inputting at least one input value as input data into the neural network. The at least one input value can be transmitted to at least one input node of the neural network, for example, by means of an input device. Alternatively, the at least one input value can be transmitted from a control device of the vehicle to at least one of the input nodes of the neural network via a data interface. The at least one input value can be used as input data for determining the clutch temperature using the neural network. Alternatively or additionally, the at least one input value can be used as input data for training the neural network.
[0022] In this embodiment, the clutch temperature is determined based on the temporal course of the input data. Temporal changes in the input data can therefore also be taken into account by the neural network. For example, a cooling phase or a heating phase of the vehicle clutch can be detected and taken into account when determining the clutch temperature. Furthermore, for example, situations in which the clutch temperature exceeds a limit value or deviates from the normal behavior of the clutch temperature during the heating or cooling phase can be taken into account. The neural network can therefore also determine the clutch temperature more accurately. Furthermore, the temporal course of the input data is not determined by the neural network itself, but is only forwarded to it in the form of processed input values. This reduces the computational effort required to determine the clutch temperature. As a result, the structure of the neural network can be constructed to be less complex. As a result, the storage bits required for the neural network can also be reduced.
[0023] According to one embodiment, at least one input value is determined in a predetermined time interval. For example, at least one input value can be determined in a time interval of 1 second, 100 ms or 10 ms. Thus, the determination of the input value can be matched to the time interval in which the neural network determined the clutch temperature. Thus, the neural network is supplied with a current input value, i.e., determined within the same time interval in which the clutch temperature was determined by the neural network. Thus, the neural network can determine the clutch temperature more accurately. Since the determination of the input value generates less computational burden, efficient computation with high performance can be achieved by decoupling the implementation of the determination of the input value from the implementation of the neural network.
[0024] In a second aspect, the invention relates to a method for training a neural network, the neural network being configured to determine a clutch temperature of a vehicle clutch. The method comprises the following steps: providing at least one value of an operating parameter of the vehicle clutch as input data; providing a clutch temperature value as output data; training the neural network using the input data and the output data in order to learn a relationship between the input data and the clutch temperature.
[0025] The value of at least one operating parameter of the vehicle clutch and the value of the clutch temperature can be detected by means of a device designed for this purpose for electrical signal processing. The detected values can be provided to an input device for inputting input data into the neural network. Alternatively, the detected values can be transmitted from the device designed for this purpose to an input node of the neural network via a data interface. Alternatively or additionally, the input data or output data can be present as synthetic data, which can be generated by converting the detected or determined data.
[0026] According to one specific embodiment, the neural network is also trained with the aid of the at least one input value determined, which represents the power supplied to the vehicle clutch. The at least one input value is determined based on processing of temporally successive values of the power supplied to the vehicle clutch.
[0027] According to one embodiment, the neural network according to the method of the first aspect is trained according to the method of the second aspect. Thus, the technical effects and advantages of the two aspects can be combined.
[0028] According to a further embodiment, at least one operating parameter of the vehicle clutch is selected from at least one of the following: the torque of the drive axle of the vehicle engine in mechanical connection with the vehicle clutch; the speed difference between two rotating clutch elements of the vehicle clutch; the speed of the rotating clutch element of the vehicle clutch; the mechanical pressure acting on the clutch element of the vehicle clutch; the current intensity of the current flowing through the clutch element of the vehicle clutch; and the oil sump temperature of the vehicle transmission at the beginning of the switching process of the vehicle clutch. The clutch elements of the vehicle clutch can be, for example, clutch plates of the vehicle clutch, which can be connected to transmit force from the engine to the drive axle. The above parameters can all represent the power delivered to the vehicle clutch. Alternatively or additionally, the above parameters can all represent the clutch temperature of the vehicle clutch. By using at least one of these parameters, it is thus possible to facilitate the determination of the clutch temperature.
[0029] In a third aspect, the present invention relates to a control device for determining a clutch temperature of a vehicle clutch. The control device comprises a computer-readable storage medium on which a neural network for determining the clutch temperature is stored. In addition, the control device comprises an identification device for identifying a switching process of the vehicle clutch and an activation device for activating the neural network in dependence on the identification of the switching process. The control device further comprises an input device for inputting input data into the neural network, wherein the input data comprises at least one value of at least one operating parameter of the vehicle clutch. The control device further comprises an output device for outputting the clutch temperature determined by the neural network and a deactivation device for deactivating the neural network in dependence on the determined clutch temperature.
[0030] According to one embodiment, the clutch temperature is determined by a neural network according to the method of the first aspect. Furthermore, the neural network has been trained according to the method of the second aspect.
[0031] The above-mentioned device of the control device according to the third aspect can be configured to receive, process and forward electrical signals. The above-mentioned device of the control device according to the third aspect can be configured to perform the method according to the first aspect and / or the second aspect. Similarly, the method according to the first aspect or the second aspect can be performed by the control device according to the third aspect. Therefore, the embodiments, technical effects and advantages explained for the first or second aspect are also applicable to the control device according to the third aspect. BRIEF DESCRIPTION OF THE DRAWINGS
[0032] Figure 1 A flowchart diagram showing the steps of a method for determining a clutch temperature of a vehicle clutch by means of a neural network according to one embodiment of the present invention;
[0033] Figure 2 A flowchart diagram showing steps of a method for training a neural network according to a further embodiment of the present invention, the neural network being configured for determining a clutch temperature of a vehicle clutch;
[0034] Figure 3 A control device for determining a clutch temperature of a vehicle clutch using a neural network according to a further specific embodiment of the present invention is schematically shown. DETAILED DESCRIPTION
[0035] Figure 1 A flow chart diagram shows the steps of a method for determining a clutch temperature of a vehicle clutch by means of a neural network according to one specific embodiment of the present invention.
[0036] In a first determination step BS1, a switching process of the vehicle clutch is identified. Figure 1In an embodiment of the present invention, the switching process is identified based on a change of the target gear of the vehicle clutch.
[0037] In a second determination step BS2, the neural network is activated for determining the clutch temperature. The activation is dependent on the detection of a switching process. Figure 1 In an embodiment of the present invention, the activation is performed based on an activation signal, which is generated by the activation device when a switching process is detected.
[0038] In a third determination step BS3, at least one value of at least one operating parameter of the vehicle clutch is input as input data into the neural network. The at least one value of the at least one operating parameter is transmitted to at least one input node of the neural network.
[0039] In a fourth determination step BS4, the clutch temperature is determined by the neural network. The clutch temperature is determined based on the input data and the relationship between the input data and the clutch temperature learned by the neural network.
[0040] In order to determine the clutch temperature, the input data transmitted to the input nodes of the neural network are transmitted to the intermediate nodes of the neural network via the connections mathematically weighted by the neural network. The transmitted input data are processed at the intermediate nodes and transmitted to the output nodes of the neural network via the connections mathematically weighted by the neural network. During a training process of the neural network preceding the determination method in time, the mathematical weights of the respective connections are adjusted by the neural network based on the training data set. At the end of the method, the value of the clutch temperature will be present at at least one output node of the neural network.
[0041] In the fifth determination step BS5, the determined clutch temperature is output from the neural network. Figure 1 In an embodiment of the present invention, the clutch temperature is output to an output device.
[0042] In a sixth determination step BS6, the neural network for determining the clutch temperature is deactivated. The deactivation is performed in dependence on the determined clutch temperature. Figure 1 In an embodiment of the present invention, the neural network is deactivated when the determined clutch temperature is below a defined threshold.
[0043] The neural network can be activated as required depending on the detection of a switching process. Furthermore, the neural network can also be deactivated as required depending on the determined clutch temperature. The determination of the clutch temperature by means of the neural network can thus be matched to a predetermined computing capacity of the control device at which the determination by means of the neural network takes place.
[0044] Figure 2 A flow chart diagram shows steps of a method for training a neural network according to a further specific embodiment of the present invention, the neural network being designed to determine a clutch temperature of a vehicle clutch.
[0045] In a first training step TS1 , a value of at least one operating parameter of a vehicle clutch is provided as input data. This value represents the clutch temperature of the vehicle clutch and is transmitted to at least one input node of the neural network.
[0046] In a second training step TS2, values of the clutch temperature are provided as output data. These values are measured values of the clutch temperature. In an embodiment not shown, the values of the clutch temperature are synthetically generated. These values are transmitted to at least one output node of the neural network.
[0047] In a third training step TS3 , the neural network is trained using the input data and the output data in order to learn the relationship between the input data and the clutch temperature.
[0048] Used according to Figure 1 Embodiments determine that the neural network of the clutch temperature can be based on Figure 2 The method of implementing the method is trained.
[0049] Figure 3 A control device 10 for determining a clutch temperature of a vehicle clutch (not shown) using a neural network 12 is schematically shown according to a further specific embodiment of the present invention.
[0050] The control device 10 comprises a computer-readable storage medium 14 on which the neural network 12 is stored. The control device 10 also comprises an input device 16 which is designed to receive values 18a, 18b, 18c of operating parameters of the vehicle clutch. The input device 16 transmits the values 18a, 18b, 18c of the operating parameters of the vehicle clutch as input data 20a to the neural network 12.
[0051] The control device 10 comprises a recognition device 22 which recognizes a switching process of a vehicle clutch based on a switching signal 24. Depending on the recognition of the switching process, the recognition device 22 generates a recognition signal 26 and transmits the recognition signal to an activation device 28. The activation device 28 generates an activation signal 30 depending on the receipt of the recognition signal 26 and transmits the activation signal to the neural network 12. The neural network 12 is activated depending on the receipt of the activation signal 30.
[0052] The control device 10 further comprises a detection device 32a, 32b, which detects temporally successive values 34a, 34b or 34c, 34d of the power supplied to the vehicle clutch. The detection devices 32a, 32b are low-pass filters, and the values 34a, 34b, 34c, 34d of the power supplied to the vehicle clutch are the switching power supplied to the vehicle clutch during the switching process or gear shift. The detected values 34a, 34b, 34c and 34d of the switching power are transmitted from the detection devices 32a, 32b to a determination device 36. The determination device 36 determines at least one average value of the switching power by processing the temporally successive values 34a, 34b, 34c, 34d. The at least one average value of the switching power is transmitted by the determination device 36 as an input value 38 to the input device 16. The input device 16 transmits the input value 38 as input data 20b to the neural network 12.
[0053] The clutch temperature KT of the vehicle clutch is determined by the neural network 12 from the input data 20a, 20b. The determined clutch temperature KT is transmitted from the neural network 12 to the output device 40. The output device 40 transmits the determined clutch temperature KT to the deactivation device 42. Depending on the determined clutch temperature KT, the deactivation device 42 generates a deactivation signal 44, by which the neural network 12 can be deactivated. Figure 3 In the exemplary embodiment of FIG. 4 , a deactivation signal 44 is generated for the neural network 12 when the determined clutch temperature KT is below a predetermined threshold value.
[0054] Reference numerals list
[0055] 10 Control Device
[0056] 12 Neural Networks
[0057] 14 Storage Media
[0058] 16 Input Devices
[0059] 18a, 18b, 18c Values of operating parameters
[0060] 20a, 20b Input data
[0061] 22 Identify device
[0062] 24 Switching signal
[0063] 26 Identification Signals
[0064] 28 Activate Device
[0065] 30 Activation signal
[0066] 32a, 32b Testing equipment
[0067] 34a, 34b, 34c, 34d Time-sequential values of the power supplied to the vehicle clutch
[0068] 36 Determine the device
[0069] 38 Input value
[0070] 40 Output Devices
[0071] 42 Deactivate device
[0072] 44 Deactivate signal
[0073] KT Clutch temperature
[0074] BS1 First Determination Step
[0075] BS2 Second determination step
[0076] BS3 Third Determination Step
[0077] BS4 Fourth Determination Step
[0078] BS5 Fifth Determination Step
[0079] BS6 Sixth Determination Step
[0080] TS1 First training step
[0081] TS2 Second training step
[0082] TS3 Third training step
Claims
1. A method for determining a clutch temperature of a vehicle clutch by means of a neural network (12), wherein: The method comprises the following steps: - identifying (BS1) a shifting process (24) of the vehicle clutch; - activating (BS2) the neural network (12) for determining the clutch temperature as a function of the detection of a switching process (24); - inputting (BS3) at least one value (18a, 18b, 18c) of at least one operating parameter of the vehicle clutch as input data (20a) into the neural network (12); - determining (BS4) a clutch temperature (KT) by means of the neural network (12) based on the input data (20a) and the relationship between the at least one operating parameter and the clutch temperature (KT) learned by the neural network (12); - output (BS5) the clutch temperature (KT) determined by said neural network (12); and - Deactivating (BS6) a neural network (12) for determining the clutch temperature as a function of the determined clutch temperature (KT).
2. The method according to claim 1, wherein If the determined clutch temperature (KT) is below a defined threshold value, the neural network (12) is deactivated.
3. The method according to claim 1 or 2, wherein: The clutch temperature is determined by the neural network (12) within a predetermined time interval.
4. The method according to claim 3, wherein: The time interval is predetermined as a function of the determined clutch temperature.
5. The method according to any one of claims 1 to 4, wherein: The method further comprises the following steps: - determining at least one input value (38) representing the power supplied to the vehicle clutch, wherein the at least one input value (38) is determined based on processing of temporally consecutive values (32a, 32b, 32c, 32d) of the power supplied to the vehicle clutch; and - inputting the at least one input value (38) as input data (20b) into the neural network (12).
6. The method according to claim 5, wherein: The at least one input value (38) is determined within a predetermined time interval.
7. A method for training a neural network (12) which is designed to determine a clutch temperature (KT) of a vehicle clutch, wherein: The method comprises the following steps: - providing (TS1) at least one value (18a, 18b, 18c) of an operating parameter of the vehicle clutch as input data (20a); - provide as output data the value of the (TS2) clutch temperature; and - training (TS3) the neural network (12) using the input data (20a) and the output data in order to learn the relationship between the input data and the clutch temperature.
8. The method according to claim 7, wherein: The neural network (12) is also trained with the aid of at least one determined input value (38), which represents the power supplied to the vehicle clutch, wherein the at least one input value (38) is determined based on processing of temporally successive values (32a, 32b, 32c, 32d) of the power supplied to the vehicle clutch.
9. The method according to any one of claims 1 to 6, wherein: The neural network (12) has The training was performed according to the method for training a neural network according to any one of claims 7 or 8.
10. A method according to any one of the preceding claims, wherein: At least one operating parameter of the vehicle clutch is selected from at least one of the following: the torque of the drive axle of the vehicle engine that is mechanically connected to the vehicle clutch; the speed difference between the two rotating clutch elements of the vehicle clutch; the speed of the rotating clutch element of the vehicle clutch; the mechanical pressure acting on the clutch element of the vehicle clutch; the current intensity of the current flowing through the clutch element of the vehicle clutch; and the oil sump temperature of the vehicle transmission when the switching process of the vehicle clutch begins.
11. A control device (10) for determining a clutch temperature of a vehicle clutch, the control device comprising: - a computer-readable storage medium (14) on which a neural network (12) for determining the clutch temperature is stored; - a recognition device (22) for recognizing a switching process (24) of the vehicle clutch; - an activation device (28) for activating (30) the neural network (12) as a function of detecting a switching process (24); - an input device (16) for inputting input data (20a, 20b) into the neural network (12), wherein the input data (20a, 20b) comprises at least one value (18a, 18b, 18c) of at least one operating parameter of the vehicle clutch; - an output device (40) for outputting the clutch temperature (KT) determined by the neural network (12); and - a deactivation device (42) for deactivating (44) the neural network (12) as a function of the determined clutch temperature (KT).
12. The control device (10) according to claim 11, wherein: The clutch temperature (KT) is determined by the neural network (12) according to the method of any one of claims 1 to 6; and wherein the neural network (12) has The training was performed according to the method for training a neural network according to any one of claims 7 or 8.