Method and apparatus for determining a residual service life based on a predictive diagnosis of components of an electric drive system using artificial intelligence techniques

Patent metadata

Field	Value
Publication number	US-12469341-B2
Application number	US-202117486093-A
Country	US
Kind code	B2
Filing date	Sep 27, 2021
Priority date	Sep 29, 2020
Publication date	Nov 11, 2025
Grant date	Nov 11, 2025

Abstract

A computer-implemented method for providing a residual service life based on a diagnosis of components of an electric drive system in a vehicle, includes recording distributions of a plurality of operating parameters comprising at least one sensor parameter and/or at least one control parameter, in the vehicle; using a plurality of diagnostic models for a plurality of fault types, each of which is configured to detect a specific fault type in one of the components based on at least some of the plurality of operating parameters and to signal corresponding fault information associated with the fault type; determining the residual service life using a residual usage model depending on the signaled corresponding fault information, the residual usage model configured to indicate the residual service life depending on the corresponding fault information from the plurality of diagnostic models; and signaling the residual service life.

First claim

What is claimed is:

1 . A method for providing a residual service life based on a diagnosis of components of an electric drive system in a vehicle, comprising:
recording a plurality of operating parameters, the plurality of operating parameters including at least one of (i) at least one sensor parameter measured by at least one sensor in the vehicle and (ii) at least one control parameter that is output to control one of the components of the electric drive system in the vehicle; determining distributions of the plurality of operating parameters; determining a plurality of fault information using a plurality of diagnostic models for a plurality of fault types, each of which is configured to detect a specific fault type in a specific one of the components of the electric drive system based on the distributions of at least some of the operating parameters of the plurality of operating parameters and provide corresponding fault information of the plurality of fault information associated with the specific fault type, the plurality of diagnostic models including at least one anomaly detection model configured to detect an anomaly in a behavior of the electric drive system based on the distributions of the at least some operating parameters of the plurality of operating parameters and to provide a corresponding degree of anomaly as the corresponding fault information, the plurality of diagnostic models including at least one physical diagnostic model configured to detect the fault in the specific one of the components of the electric drive system based on at least one of correlated distributions and redundant distributions of at least some of the plurality of operating parameters and to provide the corresponding fault information; determining the residual service life using a residual usage model depending on the plurality of fault information, the residual usage model configured to indicate the residual service life depending on the plurality of fault information, including depending on the corresponding degree of anomaly; and signaling the determined residual service life.

2 . The method according to claim 1 , wherein the plurality of diagnostic models comprise one or more data-based fault classification models, each of which is configured to detect the fault in the specific one of the components of the electric drive system based on a profile of the at least some operating parameters of the plurality of operating parameters and to provide the corresponding fault information.

3 . The method according to claim 1 , wherein the at least one anomaly detection models include an autoencoder configured to reduce a feature space of the plurality of operating parameters and reconstruct the plurality of operating parameters.

4 . The method according to claim 3 , wherein the corresponding degree of anomaly is determined depending on a deviation between the plurality of operating parameters and the reconstructed plurality of operating parameters.

5 . The method according to claim 1 , wherein:
the corresponding fault information is assigned to a critical fault or a non-critical fault in a relevant component of the electric drive system, and when the critical fault is assigned an end of the service life is signaled.

6 . The method according to claim 5 , wherein:
when the non-critical fault is assigned, timing information is assigned to the non-critical fault, which specifies when the non-critical fault occurred, the timing information is used for determining the residual service life, and the timing information is supplied to the residual usage model as an input variable.

7 . The method according to claim 5 , wherein:
the corresponding fault information is continuously transmitted to a central unit of one or a plurality of motor vehicles, and when the critical fault is assigned that indicates the end of the service life, one or more training datasets are generated which assigns the corresponding fault information of the plurality of diagnostic models to the residual service life.

8 . The method according to claim 7 , wherein:
the residual usage model is data-based and implemented as a neural network, and the residual usage model is retrained or updated with the one or more training datasets.

9 . The method according to claim 7 , wherein the residual usage model comprises a clustering procedure, which is based on a nearest neighbor approach and uses the one or more training datasets to determine the residual service life from currently supplied corresponding fault information.

10 . The method according to claim 1 , wherein a computer program product comprises commands which, during execution of the computer program product by least one data processing device, causes said at least one data processing device to execute the method.

11 . The method according to claim 10 , wherein the computer program product is stored on a non-transitory computer-readable storage medium.

12 . An apparatus for providing a residual service life based on a diagnosis of components of an electric drive system in a vehicle, comprising:
at least one data processing device configured to:
record a plurality of operating parameters, the plurality of operating parameters including at least one of (i) at least one sensor parameter measured by at least one sensor in the vehicle and (ii) at least one control parameter that is output to control one of the components of the electric drive system, in the vehicle;
determine distributions of the plurality of operating parameters;
determine a plurality of fault information using a plurality of diagnostic models for a plurality of fault types, each of which is configured to detect a specific fault type in a specific one of the components of the electric drive system based on the distributions of at least some operating parameters of the plurality of operating parameters and provide corresponding fault information of the plurality of fault information associated with the specific fault type, the plurality of diagnostic models including at least one anomaly detection model configured to detect an anomaly in a behavior of the electric drive system based on the distributions of the at least some operating parameters of the plurality of operating parameters and to provide a corresponding degree of anomaly as the corresponding fault information, the plurality of diagnostic models including at least one physical diagnostic model configured to detect the fault in the specific one of the components of the electric drive system based on at least one of correlated distributions and redundant distributions of at least some of the plurality of operating parameters and to provide the corresponding fault information;
determine the residual service life using a residual usage model depending on the plurality of fault information, the residual usage model configured to indicate the residual service life depending on the plurality of fault information, including depending on the corresponding degree of anomaly; and
signaling the residual service life.

Assignees

• Bosch Gmbh Robert

Inventors

• Simonis Christian
• Woll Christoph
• Gorelik Kirill
• Shen Tunan

Classifications

• G07C5/0808

  Diagnosing performance data (testing of vehicles G01M17/00; testing of electrical installation on vehicles G01R31/005) · CPC title

• G06N3/02

  Neural networks · CPC title

• G06N3/0455

  Auto-encoder networks; Encoder-decoder networks · CPC title

• G06N3/09

  Supervised learning · CPC title

• G06N3/045

  Combinations of networks · CPC title