DE102022124901B3 - Method for operating a vehicle device and vehicle device - Google Patents

Abstract

The invention relates to a method (10) for operating a vehicle device (12) which is dependent on at least one calculation value (16) and comprises the following steps: determining the calculation value (16) with an analytical relationship (18) depending on at least one first parameter (a) and second parameters (b) of the vehicle device (12) and an input value (TR) assigned to the vehicle device (12), calculation of the first parameter (a) by a first neural network (20), calculation of the second parameter (b) through a second neural network (22). The invention further relates to a vehicle device (12).

Description

The invention relates to a method for operating a vehicle device according to claim 1. The invention further relates to a vehicle device.

DE 10 2020 214 234 A1 discloses a method for determining hotspot temperatures of a rotor and a stator.

DE 10 2021 214 517 A1 discloses a method for determining a temperature in a component of an electric motor for an electric drive system using machine learning methods.

In WO 2020 147 870 A1 describes a method for calculating the temperature of a vehicle device, especially a drive unit. The temperature at the respective temperature detection area is calculated using a temperature model.

The object of the present invention is to operate the vehicle device more efficiently and efficiently. In particular, a calculation value, for example the temperature of the vehicle device, should be calculated more accurately and quickly.

At least one of these tasks is achieved by a method for operating a vehicle device having the features according to claim 1. As a result, the calculation value determined in this way can enable a more precise and reliable assessment of the behavior of the vehicle device. The calculation value can be calculated more accurately. This allows the vehicle device to be operated more efficiently and powerfully.

The vehicle device can be assigned to the vehicle. The vehicle device can be an electric motor, power electronics, a thermal management system, a battery and/or a transmission. The vehicle can be an electric vehicle.

The calculation value can change over time. The calculation value can be determined at intervals. The time intervals can be predetermined time intervals. The time intervals can be constant or variable. The time intervals can be adjustable depending on environmental conditions or measured values.

The first parameter can be a variable. The second parameter can be a variable.

By using the first and second neural networks, the first and second parameters can be calculated even with unknown input variables. The first and/or second neural network can have at least one input layer, one or more hidden layers and an output layer.

The first and/or second neural network may be a feed forward neural network (FNN), a recurrent neural network (RNN) or a convolutional neural network (CNN).

In a preferred embodiment of the invention, it is advantageous if the input value is a measured and/or calculated physical quantity of the vehicle device. The input value can be a determined calculation value from a previous determination of the calculation value.

In a special embodiment of the invention, it is advantageous if the first and second neural networks are trained separately from one another. This can reduce the complexity and effort required to train the neural networks. The first and second neural networks can be trained dependently or independently of one another. The first neural network can be trained with training data that, in particular, completely deviates from the training data of the second neural network. The training data of the first and second neural networks can at least partially overlap.

In a special embodiment of the invention, it is advantageous if the determined calculation value forms the input value of a further determination of the calculation value following the determination. The calculation value can be dynamic in time and can therefore be calculated depending on time.

The vehicle device is an electric motor of a powertrain of the vehicle, and the calculation value is a temperature change of a temperature of the electric motor.

According to the invention it is provided that the analytical connection can be represented by an nth order polynomial. The analytical context may include addition, subtraction, multiplication and/or division. Often it is not possible to specify exact analytical relationships for a vehicle device or several vehicle devices, for example because the systems are too complex. One possibility is to reduce complexity through simplifying assumptions, aggregations of parts/components, or similar. This approach provides approximate analytical connections. Even for this kind of analytical The procedure described here can be used in this context. In this context, the term “system” can be understood to mean parts, groups of components, components and/or individual parts of the vehicle device.

According to the invention it is provided that the number of parameters of the analytical relationship including the first and second parameters is equal to n+1. Preferably, each parameter can be calculated by its own neural network, including the first and second neural networks. This can reduce the complexity of the neural networks.

According to the invention it is provided that the number of neural networks used to determine the calculation value and including the first and second neural networks is equal to n+1. Also, including the first and second neural networks, fewer than n+1 or n neural networks can be used. For example, one of the parameters of the analytical relationship can be calculated conventionally. At least one of the parameters of the analytical context can be fixed.

In an advantageous embodiment of the invention it is provided that the calculation value indicates a temperature change, the first parameter indicates a heat input, the second parameter indicates a heat dissipation and/or the input value indicates a temperature. The temperature change can be the temperature of an electric motor. The heat input can arise from electrical energy. The heat dissipation can be conductive, radiative and/or convective.

Furthermore, at least one of the previously specified tasks is solved by a vehicle device with the features according to claim 6.

In a special embodiment of the invention, it is advantageous if the vehicle device is a drive train device of a drive train of the vehicle. The drive train device can be set up to move the vehicle. The powertrain device can generate drive power to move the vehicle.

Further advantages and advantageous refinements of the invention result from the description of the figures and the illustration.

Character description

• 1 zeigt eine schematische Ansicht eines Verfahrens zum Betrieb einer Fahrzeugvorrichtung in einer speziellen Ausführungsform der Erfindung. Das Verfahren 10 zum Betrieb einer Fahrzeugvorrichtung 12 wird beispielsweise im Betrieb eines die Fahrzeugvorrichtung 12 aufweisenden Fahrzeugs angewendet. Die Fahrzeugvorrichtung 12 kann eine Antriebsstrangvorrichtung 14 des Fahrzeugs, speziell ein Elektromotor des Fahrzeugs sein.

The invention is described in detail below with reference to the figure. It shows:

• 1 shows a schematic view of a method for operating a vehicle device in a special embodiment of the invention. The method 10 for operating a vehicle device 12 is used, for example, when operating a vehicle having the vehicle device 12. The vehicle device 12 may be a powertrain device 14 of the vehicle, specifically an electric motor of the vehicle.

The operation of the vehicle device 12 takes place depending on a calculation value 16. The calculation value 16 can be a temperature change ΔT of a temperature T of the electric motor. The electric motor can preferably be operated depending on the temperature change ΔT and thus on the temperature T calculated depending on the temperature change ΔT.

The calculation value 16 is calculated by an analytical context 18 depending on at least a first parameter a and a second parameter b and an input value T _R. For example, the analytical connection 18 can be represented by an nth order polynomial, in particular first order with n = 1, as follows: $$\Delta T=a{T}_R+b$$

einer zeitlich der Ermittlung nachfolgenden weiteren Ermittlung des Berechnungswerts 16 wie folgt bilden: $$T_R^{t+1}=T_R+\Delta T$$

The number of parameters of the analytical relationship 18 is therefore equal to n + 1. The determined calculation value 16 can be the input value $T_R^{t+1}$

a further determination of the calculation value 16 following the determination as follows: $$T_R^{t+1}=T_R+\Delta T$$

mit dem Wärmeübergangskoeffizient α, der in den Wärmeübergang einbezogenen Kontaktfläche A und der spezifischen Wärmekapazität C_T der betroffenen Bauteile.The first parameter a can characterize a heat input P _in and the second parameter b can characterize a heat dissipation P _out . The first parameter a can be structured as follows depending on input variables: $$a=\frac{\alpha A}{C_T}$$

with the heat transfer coefficient α, the contact area A included in the heat transfer and the specific heat capacity C _T of the affected components.

mit dem elektrischen Widerstand R. Dabei wurde die enstehende Wärmeleistung wie folgt angenommen: $$P_{in}=I^{2}R$$

mit dem elektrischen Strom I.The second parameter b can be structured as follows depending on input variables: $$b=\frac{R}{C_T}$$

with the electrical resistance R. The resulting thermal power was assumed as follows: $$P_{in}=I^{2}R$$

with the electric current I.

The first parameter a is calculated by a first neural network 20 and the second parameter b by a second neural network 22. The first and second neural networks 20, 22 are preferably trained independently of one another. The number of neural networks used to determine the calculation value 16 is equal to n + 1. The first neural network 20 is trained with training data a ₁ to a _m and the second neural network 22 is trained with training data b ₁ to b _k .

Reference symbol list

Pin code

Heat input

Pout

Heat dissipation

TR

Input value

ΔT

Temperature change

a

first parameter

b

second parameter

10

Proceedings

12

Vehicle device

14

Powertrain device

16

Calculation value

18

analytical context

20

first neural network

22

second neural network

Claims (6)

Method (10) for operating a vehicle device (12) which takes place depending on at least one calculation value (16) and comprises the following steps: determining the calculation value (16) with an analytical relationship (18) depending on at least one first parameter (a) and second parameter (b) of the vehicle device (12) and an input value (T _R ) assigned to the vehicle device (12), calculation of the first parameter (a) by a first neural network (20), calculation of the second parameter (b) by a second neural network (22), wherein the vehicle device (12) is an electric motor of a drive train (14) of the vehicle and the calculation value (16) is a temperature change (ΔT) of a temperature of the electric motor, the analytical relationship (18) being represented by a polynomial n -th order can be represented. wherein the number of parameters of the analytical context (18) including the first and second parameters (a, b) is equal to n+1, characterized in that the number of the first and second neural networks used to determine the calculation value (16). (20, 22) including neural networks is equal to n+1. Method (10) for operating a vehicle device (12). Claim 1 , characterized in that the input value ( _TR ) is a measured and/or calculated physical quantity of the vehicle device (12). Method (10) for operating a vehicle device (12). Claim 1 or 2 , characterized in that the first and second neural networks (20, 22) are trained separately from each other. Method (10) for operating a vehicle device (12) according to one of the preceding claims, characterized in that the determined calculation value (16) forms the input value (T _R ) of a further determination of the calculation value (16) following the determination. Method (10) for operating a vehicle device (12) according to one of the preceding claims, characterized in that the calculation value (16) is a temperature change (ΔT), the first parameter (a) is a heat input (P _in ), the second parameter (b ) a heat dissipation (P _out ) and / or the input value (T _R ) indicates a temperature (T). Vehicle device (12) which is operated according to a method (10) of the preceding claims.

Images