Method for simulating the power requirements of electrical components of a vehicle arranged in a low-voltage electrical system

Abstract

The invention relates to a method for simulating the power consumption of electrical components of a vehicle arranged in a low-voltage electrical system. The essential feature is that an interpolation of two or more values ​​is performed in or based on at least one consumer model representing the power consumption of electrical components arranged in a low-voltage electrical system, wherein the at least one consumer model is replaced by, or is, an artificial intelligence model, hereinafter referred to as an AI model, the AI ​​model being provided in a control unit of the vehicle and determining the power consumption of the vehicle's electrical components based on bus signals provided in the vehicle.

Description

[0001] The invention relates to a method for simulating the power consumption of electrical components of a vehicle arranged in a low-voltage electrical system according to the preamble of claim 1 and to a method for simulating the power consumption of electrical components of a vehicle arranged in a low-voltage electrical system according to the preamble of claim 2.

[0002] A method for simulating the power consumption of electrical components arranged in a vehicle's low-voltage electrical system is described in DE 10 2021 006 280 A1. A disadvantage is that the models used only provide an inadequate prediction of the power consumption of the components of a vehicle's low-voltage electrical system.

[0003] The object of the invention is therefore to provide an improved or at least a different embodiment of a method for simulating the power consumption of electrical components of a vehicle arranged in a low-voltage electrical system. In particular, an attempt is made to improve the qualitative prediction of the power consumption of these components.

[0004] In the present invention, this object is achieved in particular by the subject matter of the independent claims. Advantageous embodiments are the subject matter of the dependent claims and the description.

[0005] The basic idea of ​​the invention is to replace at least one consumer model used in a method for simulating the power consumption of electrical components of a vehicle arranged in a low-voltage electrical system with an artificial intelligence model, hereinafter referred to as an AI model. The replacement can be achieved by linking.

[0006] According to a first method according to the invention for simulating the power demand of electrical components of a vehicle arranged in a low-voltage electrical system, it is proposed that an interpolation of two or more values ​​be performed in or based on at least one consumer model representing the power demand of electrical components arranged in a low-voltage electrical system. Furthermore, within the scope of the proposed method, it is provided that the at least one consumer model is or is replaced by an artificial intelligence model, hereinafter referred to as the AI ​​model.

[0007] According to a second method according to the invention for simulating the power demand of electrical components of a vehicle arranged in a low-voltage electrical system, it is proposed that at least one consumer model, which represents the power demand of electrical components arranged in a low-voltage electrical system and which is used in the method, is replaced by an artificial intelligence model, hereinafter referred to as an AI model.

[0008] By using an AI model instead of at least one consumer model, a qualitative improvement in the prediction of the power demand of the components of a vehicle's low-voltage electrical system can be achieved.

[0009] The AI ​​model can, for example, be provided in a vehicle control unit and determine the power requirements of the vehicle's electrical components based on bus signals provided in the vehicle.

[0010] Said at least one consumer model is expediently a representation of the behavior, in particular a behavior with regard to the power demand, of the electrical components arranged in a low-voltage electrical system. Said at least one consumer model is expediently at least one characteristic map model, more particularly at least one characteristic map model within the Matlab / Simulink software, and / or at least one Matlab / Simulink model, more particularly at least one Matlab / Simulink model within the Matlab / Simulink software.

[0011] It is useful if at least one consumer model is replaced by a link to an AI model.

[0012] Furthermore, it can be provided that the AI ​​model is formed by a decision tree model, a neural network, in particular a multi-layer perceptron and / or a recurrent neural network or a support vector machine. The AI ​​model can in particular be formed by several neural networks, in particular several multi-layer perceptrons and / or several recurrent neural networks.

[0013] It can be advantageous to train the AI ​​model based on measurement files. "Measurement files" are, in particular, measurement files, i.e., files that are empirically determined, for example, in experiments.

[0014] Furthermore, it may be expedient to provide for the AI ​​model to be trained based on learned data sets and / or an artificial change in the environment, for example, a change in temperature. In other words, the method is based on learned data sets and / or an additional artificially induced change in the environment.

[0015] It can also be planned that the method is based on data-driven models. This means that the applied AI model is used with a large amount of data, whereby the qualitative improvement in the prediction of electricity demand is greater the more extensive the training dataset used.

[0016] In particular, the method can be operated or executed based on a Matlab / Simulink model. Existing Matlab / Simulink models can be replaced by an AI model.

[0017] It is understood that the above-mentioned features can be used not only in the combination specified in each case, but also in other combinations or in isolation, without departing from the scope of the present invention. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] DE 10 2021 006 280 A1

[0002]

Claims

[1] Method for simulating the power requirements of electrical components of a vehicle arranged in a low-voltage electrical system, characterized by , that - an interpolation of two or more values ​​is carried out in or based on at least one consumer model representing the power requirements of electrical components arranged in a low-voltage vehicle electrical system, - wherein the at least one consumer model is or is replaced by an artificial intelligence model, hereinafter referred to as AI model. [2] Method for simulating the power requirements of electrical components of a vehicle arranged in a low-voltage electrical system, characterized by , that - at least one consumer model, which represents the power requirements of electrical components arranged in a low-voltage vehicle electrical system and is used in the method, is replaced by an artificial intelligence model, hereinafter referred to as the Kl model. [3] Method according to claim 1 or 2, characterized by , that - that at least one consumer model is replaced by a link to a Kl model. [4] Method according to one of the preceding claims, characterized by , that - the Kl model by a o Decision tree model, and / or o a neural network, in particular a multi-layer perceptron, a recurrent neural network or a support vector machine, is formed. [5] Method according to claim 4, characterized by , that - the AI ​​model is trained based on measurement data. [6] Method according to one of the preceding claims, characterized by , that - the AI ​​model is trained based on learned data sets and / or an artificial change in the environment. [7] Method according to one of the preceding claims, characterized by , that - the process is based on data-driven models. [8] Method according to one of the preceding claims, characterized by , that - the process is operated or executed on the basis of a Matlab / Simulink model.