Method and Device for the Robust Prediction of the Aging Behavior of an Energy Storage Unit in a Battery-Operated Machine

Patent metadata

Field	Value
Publication number	US-2022097561-A1
Application number	US-202117487895-A
Country	US
Kind code	A1
Filing date	Sep 28, 2021
Priority date	Sep 29, 2020
Publication date	Mar 31, 2022
Grant date	—

Abstract

A method for determining a predicted state of health of an electrical energy storage unit in a machine includes providing a data-based or hybrid state of health model, the data-based state of health model is trained, depending on operating variables of the electrical energy storage unit and/or operating features derived from the operating variables, to indicate a state of health and to indicate a model uncertainty of the indicated state of health, ascertaining a state of health characteristic and the associated model uncertainty of the energy storage unit based on the operating variables using the state of health model, and generating at least one random constructed state of health characteristic candidate that corresponds to constructed state of health characteristics within characteristic of confidence intervals, the characteristic of confidence intervals defined by the model uncertainties of the states of health of the ascertained state of health characteristic.

First claim

What is claimed is:

1 . A method for determining a predicted state of health of an electrical energy storage unit in a machine, comprising:
providing a data-based or hybrid state of health model, the data-based state of health model is trained, depending on operating variables of the electrical energy storage unit and/or operating features derived from the operating variables, to indicate a state of health and to indicate a model uncertainty of the indicated state of health; ascertaining a state of health characteristic and the associated model uncertainty of the energy storage unit based on the operating variables using the state of health model; generating at least one random constructed state of health characteristic candidate that corresponds to constructed state of health characteristics within characteristic of confidence intervals, the characteristic of confidence intervals defined by the model uncertainties of the states of health of the ascertained state of health characteristic; selecting, from a plurality of provided real state of health characteristics of real energy storage units of other machines, a number of the real state of health characteristics that are closest to the at least one random constructed state of health characteristic candidate; ascertaining a probability density of the selected number of real state of health characteristics in order to determine a characteristic of the average of the state of health as a predicted state of health characteristic; and signaling the predicted state of health characteristic.

2 . The method according to claim 1 , wherein:
the at least one random constructed state of health characteristic candidates is checked for plausibility before the at least one random constructed state of health characteristic candidate is used to ascertain a probability density function used to ascertain the probability density, and implausible state of health characteristic candidates of the at least one random constructed state of health characteristic candidate are not taken into consideration for ascertaining the probability density function.

3 . The method according to claim 2 , wherein:
generating the at least one random constructed state of health characteristic candidate and the selecting the number of the real state of health characteristics are performed multiple times, the selected real state of health characteristics are each added to an evaluation set, and the probability density function is ascertained based on the selected real state of health characteristics of the evaluation set.

4 . The method according to claim 3 , wherein:
generating the at least one random constructed state of health characteristic candidates and the selecting the number of the real state of health characteristics are performed until a convergence criterion is met, the convergence criterion depends on an overall confidence that results from a characteristic of a variance or a standard deviation of the probability density function, and the convergence criterion depends on a change in an overall confidence in successive run-throughs of the generation and the selection.

5 . The method according to claim 1 , wherein:
the selection of the number of the real state of health characteristics that are closest to the at least one random constructed state of health characteristic candidate is performed using a nearest neighbor method, a least squares method, in a multidimensional operating feature space, and/or with respect to temporal or usage-dependent characteristic of the respective state of health characteristic candidates.

6 . The method according to claim 1 , wherein:
the predicted state of health characteristic is transferred to the battery-operated machine, and the battery-operated machine is operated based on the predicted state of health characteristic.

7 . The method according to claim 1 , wherein the predicted state of health characteristic is used to ascertain a remaining service life with a statistical uncertainty quantification and to use or to signal the remaining service life in subsequent methods for outputting operating recommendations.

8 . The method according to claim 2 , wherein:
the predicted state of health characteristic is used to signal an exchange of the electrical energy storage unit upon reaching a remaining service life of zero when a prediction uncertainty falls below a predefined uncertainty threshold, and the prediction uncertainty is calculated through integration over the probability density function with respect to the remaining service life.

9 . The method according to claim 1 , wherein the probability density is ascertained by adding histograms of states of health, ascertained at each evaluation time, of the selected real state of health characteristics.

10 . The method according to claim 1 , wherein the electrical energy storage unit is used to operate a machine, a motor vehicle, a pedelec, an aircraft, a drone, a machine tool, a consumer electronics device, a cell phone, an autonomous robot, and/or a domestic appliance.

11 . The method according to claim 1 , wherein a computer program product comprises instructions that, when the computer program product is executed by at least one data processing device, cause the at least one data processing device to perform the method.

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

13 . A device for determining a predicted state of health of an electrical energy storage unit in a battery-operated machine, comprising:
at least one data processing device configured to:
provide a data-based or hybrid state of health model, the data-based state of health model is trained, depending on operating variables of the electrical energy storage unit and/or operating features derived from the operating variables, to indicate a state of health and to indicate a model uncertainty of the indicated state of health;
ascertain a state of health characteristic and the associated model uncertainty of the energy storage unit based on the operating variables using the state of health model;
generate at least one random constructed state of health characteristic candidate that corresponds to constructed state of health characteristics within characteristic of confidence intervals, the characteristic of confidence intervals defined by the model uncertainties of the states of health of the ascertained state of health characteristic;
select, from a plurality of provided real state of health characteristics of real energy storage units of other machines, a number of the real state of health characteristics that are closest to the at least one random constructed state of health characteristic candidate;
ascertain a probability density of the selected number of real state of health characteristics in order to determine a characteristic of the average of the state of health as a predicted state of health characteristic; and
signal the predicted state of health characteristic.

Assignees

• Bosch Gmbh Robert

Inventors

• Simonis Christian
• Woll Christoph

Classifications

• G06N7/01

  Probabilistic graphical models, e.g. probabilistic networks · CPC title

• H02J2105/37

  exchanging power with electric vehicles [EV] or with hybrid electric vehicles [HEV] · CPC title

• H02J7/84

  Control of state of health [SOH] · CPC title

• Y02T10/70

  Energy storage systems for electromobility, e.g. batteries · CPC title

• G01R31/392

  Determining battery ageing or deterioration, e.g. state of health · CPC title