Method, computer-implemented tool and battery management system for estimating states of health of batteries storing electrical enery and battery energy storage system

Patent metadata

Field	Value
Publication number	US-2021302502-A1
Application number	US-202117205356-A
Country	US
Kind code	A1
Filing date	Mar 18, 2021
Priority date	Mar 24, 2020
Publication date	Sep 30, 2021
Grant date	—

Abstract

In order to estimate states of health of batteries storing electrical energy significantly indicated by battery capacity and battery internal resistance, by which the estimation of states of health of batteries, deteriorated over time usage, is automated and carried out without a need of any specific test such as a capacity test and the usage of data out of normal operation, it is proposed to collect data from normal operation of a battery storing electrical energy relating to battery-internal physical properties such as a terminal direct current IDC, a terminal direct voltage UDV and a battery cell temperature T due to battery measurements and the determination or estimation, based on this data and a battery model, of a model parameter by solving an optimization-/model parameter estimation-problem and minimizing a difference between the battery model and the battery measurements.

First claim

1 . A method for estimating states-of-health of batteries storing electrical energy, wherein
a) from a battery storing electrical energy due to battery measurements of battery-internal physical properties inter alia the measurement of a terminal direct current I DC as a first physical property, a terminal direct voltage U DV as a second physical property and a battery cell temperature T as a third physical property during a recurring, continuous, time interval, at the most over a lifetime of the battery, direct current measurement data, direct voltage measurement data and temperature measurement data are collected and stored, wherein: b) Selecting a time period with tp≥n·ti and n∈N, c) executing a dynamical battery model for the time period, which relates to measurable model input sizes u, measurable model output sizes y, model states x and model parameter z and which is a time t discretized such that a model state x t+ i is defined to a first function f with x t+1 :=f tp (x t ,u t ,z) and a model output size y t is defined to a second function g with y t :=g tp (x t ,u t ,z), by accessing to direct current evaluation data including direct current measurement data of the stored direct current measurement data relating to the measurable model input sizes u, direct voltage evaluation data including direct voltage measurement data of the stored direct voltage measurement data relating to the measurable model output sizes y and temperature evaluation data including temperature measurement data of the stored temperature measurement data relating to the measurable model input sizes u, solving an optimization-/model parameter estimation-problem given by the first function f, the second function g, the model input sizes u and the model output sizes y such that the solution of the problem yields a model parameter for the time period minimizing a difference between the battery model and the battery measurements and enabling the indication of the battery-state-of-health due to being constant for the selected time period, d) evaluating or doing a state-of-health trend analysis of an evolutionary course of the model parameter being determined by solving the optimization-/model parameter estimation-problem over numerous time periods with constant or variable time durations.

2 . The method according to claim 1 , wherein
an Equivalent Circuit Model the battery model is used as, the model input sizes u depending on a “the time t”-dependent terminal direct current I DC (t) and the battery cell temperature T with u=[I DC (t), T, . . . ], the model output sizes y depending on a “the time t”-dependent terminal direct voltage U DV (t) with y=[U DV (t), . . . ], the model states x depending on a “the time t”-dependent state-of-charge SOC with x=[SOC(t), . . . ] and the model parameter z depending on an open circuit voltage U OC being dependent on a “the time t”-dependent state-of-charge SOC(t), depending on a battery-internal resistance R and depending on a battery-internal capacity C with z=[U OC (SOC(t), C, R, . . . ] and C=C1+C2; R=R 0 +R 1 +R 2 being related to and the optimization-/model parameter estimation-problem with regard to the model parameter and the time period is solved according to the formulas
Error( z )=mean[( Y tp,BMO ( g tp , U tp,BME )− Y tp,BME ) 2 ] and z tp =argmin z [Error( z )] with

Y tp,BME [ED DV,tp ]:={y 1,BME [MD DV,ti=1 : U DV,ti=1 ( t )], y 2,BME [MD D V,ti=2 : U DV,ti=2 ( t )], . . . y n,BME [MD DV,ti=n :=U DV,ti=n ( t )]}
U tp,BME [ED DC,tp , ED T,tp ]:={u 1,BME [MD D C,ti=1 :=I DC,ti=1 ( t ), MD T,ti=1 : T ti=1 ], u 2,BME [MD DC,ti=2 :=I DC,ti=2 ( t ), MD T,ti=2 :=T ti=2 ], . . . u n,BME [MD DC,ti=n :=I DC,ti=n ( t ), MD T,ti=n :=T ti=n ]}
Y tp,BMO ( g tp , U tp,BME [ED DC,tp , ED T,tp ]):={ y 1,BMO :=g ti=1 ( x 0 , u 1,BME [MD DC,ti=1 :=I D C,ti=1 ( t ) , MD T,ti=1 :=T ti=1 ], z ) y 2,BMO :g ti=2 ( x 1 , u 2,BME [MD DC,ti=2 :=I DC,ti=2 ( t ), MD T,ti=2 :=T ti=2 ], z ), . . . y n,BMO :=g ti=1 ( x n−1 , u n,BME [MD DC,ti=n :=I DC,ti=n ( t ), MD T,ti=n :=T ti−n ], z )}.

3 . The method according to claim 2 , wherein
to solve the optimization-/model parameter estimation-problem of the Equivalent Circuit Model with regard to the model parameter a parameter ambiguity due to a dependency of the model parameter z on the open circuit voltage U OC and the battery-internal capacity C for determining the “the time t”-dependent state-of-charge SOC(t) is eliminated by calibrating the open circuit voltage U OC and the state-of-charge SOC(t) as a result of considering a defined value of the capacity C, in particular a known value of the capacity C, into the solution of the optimization-/model parameter estimation-problem.

4 . The method according to claim 1 , wherein
a notification information for notifying about an increase of the battery-state-of-health more than expected or in case of battery operating safety risks is generated as result of evaluating or doing the state-of-health trend analysis of the model parameter respectively the battery-internal resistance R and/or the battery-internal capacity C.

5 . A computer-implemented tool, in particular an Application Software <App>, for estimating states-of-health of batteries storing electrical energy, wherein
a) from a battery storing electrical energy due to battery measurements of battery-internal physical properties inter alia the measurement of a terminal direct current I DC as a first physical property, a terminal direct voltage U DV as a second physical property and a battery cell temperature T as a third physical property during a recurring, in particular continuous, time interval (ti), in particular at the most over a lifetime of the battery, direct current measurement data, direct voltage measurement data and temperature measurement data are collected and stored, wherein: a non-transitory, processor-readable storage medium having processor-readable program-instructions of a program module for estimating a state-of-health of the battery stored in the non-transitory, processor-readable storage medium and a processor connected with the storage medium executing the processor-readable program-instructions of the program module to estimate the state-of-health, wherein the program module and the processor form a calculation engine such that: b) a time period with tp>n·ti and n∈N is selected, c) a dynamical battery model for the time period, which relates to measurable model input sizes u, measurable model output sizes y, model states x and model parameter z and which is a time t discretized such that a model state x t+1 is defined to a first function f with x t+1 :=f tp (x t ,u t ,z) and a model output size y t is defined to a second function g with y t :=g tp (x t ,u t ,z), is executed by accessing to, inputting into the calculation engine, direct current evaluation data including direct current measurement data of the stored direct current measurement data relating to the measurable model input sizes u, direct voltage evaluation data including direct voltage measurement data of the stored direct voltage measurement data relating to the measurable model output sizes y and temperature evaluation data including temperature measurement data of the stored temperature measurement data relating to the measurable model input sizes u, solving an optimization-/model parameter estimation-problem given by the first function f, the second function g, the model input sizes u and the model output sizes y such that the solution of the problem yields a model parameter for the time period minimizing a difference between the batt

Assignees

• Siemens Ag

Inventors

• Bamberger Joachim
• Mesanovic Amer
• Szabo Andrei

Classifications

• G01R31/392 Primary

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

• G01R31/367 Primary

  Software therefor, e.g. for battery testing using modelling or look-up tables · CPC title

• G01R31/389

  Measuring internal impedance, internal conductance or related variables · CPC title

• G01R31/3842

  combining voltage and current measurements · CPC title