SE1550770A1

SE1550770A1 - A method and a monitoring unit for monitoring a battery system

Info

Publication number: SE1550770A1
Application number: SE1550770A
Authority: SE
Inventors: Svens Pontus
Original assignee: Scania Cv Ab
Priority date: 2015-06-09
Filing date: 2015-06-09
Publication date: 2016-12-10
Also published as: WO2016200319A1; SE539562C2; DE112016002067T5

Abstract

18 ABSTRACT A method of monitoring a battery system (1) comprising aplurality of battery cell units (2) and a battery balancing module(8). The method comprises collecting measurement data for eachof the battery cell units, based thereon determining at least oneperformance related parameter for each of the battery cell units,storing data relating to said parameter in a database, and basedon said stored data, determining a change in at least one ofcapacity and impedance of the battery system over time. Thedata the at performance related parameter are collected during a balancing measurement used to determine least one process of the battery system. (Pig. 1)

Description

A method and a monitorinq unit for monitorinq a batterv system FIELD OF THE INVENTION The invention relates to a method of monitoring a battery systemaccording to the preamble of claim 1 and to a monitoring unitaccording to the independent device claim. The invention isparticularly intended for use with battery systems in electrifiedvehicles. The invention also relates to a computer program, acomputer program product, an electronic control unit, and a motor vehicle.

BACKGROUND AND PRIOR ART Electrified vehicles, both hybrid vehicles and pure electric drivevehicles, use batteries comprising a number of individual batterycell units connected in series, each comprising one or morebattery cells connected in parallel. Internal sensors are used tomonitor parameters such as temperature, current and voltagewithin the battery, and those parameters are used to determinethe present condition of the battery. Depending on theapplication, different aspects of the battery's condition are ofinterest, e.g. the ability to deliver a particular electric power or tosupport a desired load when requested, or the state of charge (SOC) of the battery.

The individual battery cell units of the battery typically have slightly different capacities and may be at different levels of state of charge. ln order to minimize wear on the battery and maximizebattery life and run time, the battery periodically needs to bebalanced. ln other words, energy is transferred to or fromindividual cell units until all cell units are at the same level of state of charge.

Today's batteries for electrified vehicles have a limited batterylife that depends on the use of the battery. lt is therefore difficultto predict when it will be necessary to replace a battery in anelectrified vehicle. ln order to avoid a situation in which a worn-out battery starts to cause problems in the vehicle, it is commonto set a shorter period for battery replacement than necessary.Due to the high costs associated with battery replacement, it ishowever desirable to be able to more accurately predict when a battery replacement will be necessary.

US 2012/0068715 discloses a battery system comprising aplurality of battery cell units and a battery balancing module forbalancing the battery. The battery system is further provided withmonitoring units arranged to monitor performance relatedparameters for individual battery cell units, such as state ofcharge, output voltage, temperature, impedance, etc. Thoseparameters are thereafter stored in a memory on the monitoring unit and can be used to estimate battery life.

SUMMARY OF THE INVENTION lt is an object of the present invention to provide an, in at least some aspect, improved solution for accurate prediction of the state-of-health of a battery system.

This object is, according to a first aspect of the invention,achieved by the initially defined method of monitoring a batterysystem. The method is characterized in that the measurementdata used to determine the at least one performance relatedparameter are co||ected during a balancing process of the batterysystem. By using measurement data co||ected during thebalancing process of the battery system, the determination of thecapacity and/or impedance of the battery system will beaccurate, since the current in the battery system is low during thebalancing process. Problems originating from short and irregularcurrent pulses of large magnitude, common during operation ofthe battery, are thereby avoided. The impedance and thecapacity can be used on their own, or preferably together, topredict the state-of-health of the battery system. When thee.g.predetermined threshold level, the battery system is no longer impedance becomes too high, when it exceeds aable to deliver sufficient power. When the capacity is reduced,the run time of the battery decreases and it needs more frequent recharging.

The method according to the invention is particularly suitable foruse in hybrid vehicles, in which the battery system is generallyused in a relatively narrow state of charge range, i.e. it is rarelyfully charged or fully discharged. Typically, the battery system ofa hybrid vehicle operates at a state of charge of 20-50% of full charge.

The determined change in capacity and/or impedance of the battery system over time may be stored in a database in a vehicle comprising the battery system, or transferred to anexternal database, e.g. at a service center. The transfer of datamay be either by continuous wireless transfer, or upon service ofthe vehicle. Since data relating to each individual cell unit arestored, it is also possible to use stored data to see if anindividual battery cell unit is impaired, and not only to monitor the overall capacity and/or impedance of the battery system.

According to an embodiment of the invention, said measurementdata are collected at least at the start and/or at the end of thebalancing process. At the start and at the end of the balancingprocess, a known resistance is switched in parallel with eachbattery cell unit and a step in voltage and current occurs. Thevoltage and current characteristics following the step may beused to accurately determine the impedance of the individualbattery cell unit. This may be done either at the start or at theend of the balancing process, or for an even more accuratedetermination, both at the start and at the end. For determinationof the capacity of the battery system, it is useful to usemeasurement data from the start and the end of the balancingprocess, and from comparison with a known discharge curve ofthe battery system determine a change in state of charge, fromwhich the capacity may be determined. embodiment of the invention, said According to another measurement data relate to at least voltage, current andtemperature. Using these measurement data, it is possible todetermine both the impedance and the capacity of the battery system with sufficient accuracy.

According to another embodiment of the invention, a learningprocess is used in the step of determining the change in capacityand/or in impedance of the battery system over time. This is veryuseful for battery types having a flat discharge curve within acertain state of charge window, for which it is generally difficultto determine the state of charge by measuring the voltage at thestart of the balancing process. By using a learning process, suchas a neural network or a support vector machine (SVM), it is possible to improve the accuracy of the determination over time.

According to another embodiment of the invention, the at leastone performance related parameter includes at least one of stateof charge of the battery cell unit and impedance of the batterycell unit. These performance related parameters are bothpossible to directly determine from the available measurementdata, such as current, temperature and voltage. The state ofcharge may preferably be used for the determination of the capacity of the battery system while the impedance of the individual battery cell units is used to obtain the overallimpedance of the battery system.According to another embodiment of the invention, said measurement data are collected both at the start and at the endof the balancing process, and a change in capacity of the batterysystem over time is determined based on stored data relating toa change in state of charge of each of the battery cell units fromthe start to the end of the balancing process. The state of chargeof each battery cell unit at the end of the balancing process isis therefore well known through the measured voltage and suitable to use as a reference point. The state of charge at the start of the balancing process may or may not be easy to defineby the measured cell voltage, depending on battery chemistry,and it may be necessary to use a learning process to moreaccurately determine the state of charge at the beginning of thebalancing process.

According to another embodiment of the invention, theimpedance of each of the battery cell units is determined basedon measurement data collected in connection with switching inand out a known resistance in parallel with each of the batterycell units. This is practical, since a step in current and voltageresults upon switching, and the quotient between the voltage andcurrent characteristics following the switch, e.g. during thesecond immediately following the switch, are suitable for use toaccurately determine the impedance of the individual battery cellunits. Since balancing resistors of the battery system areswitched in at the start and out at the end of the balancingprocess, it is suitable to determine the impedance based onmeasurement data collected at the start and at the end of the balancing process.

According to another embodiment of the invention, a change inimpedance of the battery system over time is determined basedon stored data relating to the impedance of each of the batterycell units. This is an efficient way to determine the change in impedance of the battery system over time.

According to another aspect of the present invention, the abovedefined object is achieved by a monitoring unit for monitoring a battery system as initially defined. The monitoring unit is characterized in that it is configured to use measurement datacollected during a balancing process of the battery system todetermine said at least one performance related parameter. Theadvantages of such a unit as well as preferred embodimentsthereof are apparent from the above discussion relating to theproposed method. ln other aspects, the invention also relates to a computerprogram having the features of claim 10, a computer programproduct having the features of claim 11, an electronic control unithaving the features of claim 12 and a motor vehicle according toclaims 13 and 14.

Other advantageous features as well as advantages of the present invention will appear from the following detailed descnpüon.

BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 schematically shows a monitoring unit according to anembodiment of the invention and a battery system, Fig. 2 is a flow chart showing a method according to anembodiment of the invention, Fig. 3 is a flow chart showing a method according to anotherembodiment of the invention, and Fig. 4 schematically shows an electronic control unit according to an embodiment of the invention.

DETAILED DESCRIPTION OF EMBODIMENTS OF THEINVENTION A battery system 1 is schematically shown in fig. 1. The batterysystem 1 comprises three battery cell units 2 connected inseries. Each individual battery cell unit 2 comprises two batterycells 3 connected in parallel. The battery system 1 comprises abalancing module 8, comprising a known resistance in the form ofa resistor 4 connected in parallel with each of the battery cellunits 2 via a switch 5. A switch 13 is provided for connecting anddisconnecting the battery system 1 to a load (not shown), suchas an electric motor, and/or a battery charging unit (not shown).Of course, the number of battery cells 3 within each battery cellunit 2 may vary, as well as the number of battery cell units 2. Thenumber of battery cell units 2 may be several hundred inpractical applications, such as in the battery system of a hybridvehicle. The shown battery system 1 for simplicity and ease ofillustration only comprises a limited number of individual batterycell units 2 and battery cells 3. Furthermore, individualcomponents of the battery cell unit 2 are in fig. 1 for clarity onlymarked in one of the shown battery cell units 2.

The battery system 1 is provided with a monitoring unit 10according to an embodiment of the invention. The monitoring unit10 comprises measurement means for collecting measurementdata, including voltmeters 6 for measuring the voltage acrosseach of the battery cell units 2, an ammeter 7 for measuring thecurrent through the battery system 1, and temperature sensors 9 for measuring the temperature within individual battery cell units 2. The monitoring unit 10 further comprises data storage means11 and processing means 12. ln the data storage means 11,collected measurement data and data calculated by means of theThe processing means 12 is configured to, based on stored data processing means 12 may be stored in a database.re|ating to voltage, current and temperature, determine at leastone performance related parameter for each of the battery cellunits 2, such as state of charge (SOC) and/or impedance. Theprocessing means 12 is further configured to determine a changein capacity and/or impedance of the battery system 1 over timebased on stored data re|ating to state of charge and/or impedance of individual battery cell units 2.

The monitoring unit and the method according to the invention issuitably used for monitoring a battery system in a hybrid vehicle,used for powering an electric motor. The method is performedduring balancing of the battery system 1, when the electric motor isinacüve.

Reference is now made to fig. 2 illustrating a method accordingto an embodiment of the invention, in which the capacity of thebattery system 1 is monitored. According to the shownembodiment, the method is initiated at the start of a balancingprocess of the battery system 1. ln a first step A1, measurementdata re|ating to temperature in, current through and voltageacross each individual battery cell unit 2 are collected using thetemperature sensors 9, ammeter 7, and the voltmeters 6. ln astep A2, the state of charge (SOC) for each of the individualis determined based on the collected The battery cell units 2 measurement data using the processing means 12. determined states of charge are stored in a database in the datastorage means 11 in a step A3. At least at the end of thebalancing process, steps A1-A3 are repeated. lt is also possibleto repeat steps A1-A3 several times during the balancingprocess, until the balancing process ends. A time series of datais thereby stored in the database. ln a step A4, a change in stateof charge ASOC from the beginning of the balancing process tothe end of the balancing process is calculated in the processingmeans 12. From the change ASOC of each individual battery cellunit 2, the current capacity of the battery system 1 can be calculated in a step A5.

At the end of the balancing process, the state of charge of eachbattery cell unit 2 can be easily determined from the measuredvoltage across the battery cell unit 2 by comparison with a knowndischarge curve of the battery system 1. Depending on thebattery type, the state of charge at the beginning of the balancingprocess can be more difficult to determine from the measuredvoltage. This is the case e.g. for battery types with a relativelyflat discharge curve, for which a particular voltage across thebattery cell unit 2 cannot be unambiguously associated with aparticular state of charge. A learning process, such as a neuralnetwork or a support vector machine (SVM) may therefore beused to more accurately be able to determine the state of chargeat the beginning of the balancing process. For battery types witha more steep discharge curve, it is not necessary to use alearning process. For such battery types, a particular voltageunit 2 particular state of charge. The state of charge of individual across the battery cell is closely associated with a 11 battery cell units 2 may of course also be determined in other ways known in the art.

Another embodiment of the method according to the invention isshown in fig. 3. According to this embodiment, the method isused to monitor the impedance of the battery system 1. Themethod is initiated at the start of a balancing process of thebattery system 1. At this point, the resistors 4 are switched inusing the switch 5, resu|ting in a step in current through andvoltage across each individual battery cell unit 2. Alternatively, oradditionally, the method is initiated at the end of the balancingprocess, when the resistors 4 are switched out and another stepin current and voltage occurs. ln a step B1, measurement datarelating to temperature in, current through and voltage acrosseach individual battery cell unit 2 during the time periodimmediately following the switch are collected. ln a step B2, thequotient between the voltage and current characteristicsfollowing the switch are used to determine the impedance of theindividual battery cell units 2. The determined impedance of eachindividual battery cell unit 2 is stored in a database in a step B3.Thereafter, in a step B4, the impedance of the battery system 1is determined using the stored data relating to the impedance of each individual battery cell unit 2.

Since balancing resistors 4 of the battery system are switched inat the start and out at the end of the balancing process, it issuitable to determine the impedance based on measurement datacollected at the start and at the end of the balancing process. ltis also possible to determine the impedance based on data measured only on one occasion during the balancing process. ln 12 that case, the collection of measurement data is preferably doneat the end of the balancing process, when the battery cell unit 2 is at a well-known state of charge.

Of course the method according to the embodiments shown infig. 2 and 3 may be combined, so that both the capacity and the impedance of the battery system 1 is monitored.

The determined impedance and/or capacity of the battery system1 may be stored in a central database to allow a comparison overtime. This database may be located at e.g. a service center of avehicle in which the battery system 1 is mounted. Data may betransferred either continuously using wireless transmission, orupon service of the vehicle. The monitored changes inimpedance and/or capacity over time allow a vehicle supplier topredict when the battery system of the vehicle will need to beexchanged. On one hand, this allows prevention of a situation inwhich the determined capacity of the battery system is found tobe too low or the impedance is found to be too high forsatisfactory operation of the battery. On the other hand, it alsoprevents too frequent battery system exchanges, thus reducing the total cost per travelled kilometer.

Computer program code for implementing a method according tothe invention is suitably included in a computer program which isreadable into an internal memory of a computer, such as the in-ternal memory of an electronic control unit of a motor vehicle.Such a computer program is suitably provided through a com-puter program product comprising a data storing medium read- able by an electronic control unit, which data storing medium has 13 the computer program stored thereon. Said data storing mediumis for example an optical data storing medium in the form of aCD-ROM-disc, a DVD-disc, etc., a magnetic data storing mediumin the form of a hard disc, a diskette, a tape etc., or a Flashmemory or a memory of the type ROM, PROM, EPROM orEEPROM.

Fig. 4 illustrates very schematically an electronic control unit 40comprising an execution means 41, such as a central processorunit (CPU), for executing a computer program. The executionmeans 41 communicates with a memory 42, for example of thetype RAM, through a data bus 43. The control unit 40 comprisesalso a non-transitory data storing medium 44, for example in theform of a Flash memory or a memory of the type ROM, PROM,EPROM or EEPROM. The execution means 41 communicateswith the data storing medium 44 through the data bus 43. Acomputer program comprising computer program code forimplementing a method according to the invention is stored on the data storing medium 44.

The invention is of course not in any way restricted to the em-bodiments described above, but many possibilities to modifica-tions thereof would be apparent to a person with skill in the artwithout departing from the scope of the invention as defined in the appended claims.

Claims

1. A method of monitoring a battery system (1) comprising aplurality of battery cell units (2) and a battery balancing module(8), the method comprising: - collecting measurement data for each of the battery cellunits (2), - based on said measurement data, determining at least oneperformance related parameter for each of the battery cellunits (2), - storing data relating to said at least one performancerelated parameter in a database, - based on said stored data, determining a change in at leastone of capacity and impedance of the battery system (1)over time, characterized in that said measurement data used to determine the at least oneperformance related parameter are collected during a balancingprocess of the battery system (1 ). claim 1, wherein said

2. The method measurement data are collected at least at the start and/or at the according toend of the balancing process.or 2, wherein said

3. The method according to claim 1 measurement data relate to at least voltage, current and temperature.

4. The method according to any one of the preceding claims, wherein a learning process is used in the step of determining the change in capacity and/or in impedance of the battery system (1) over time.

5. The method according to any of the preceding claims,wherein the at least one performance related parameter includesat least one of state of charge of the battery cell unit (2) and impedance of the battery cell unit (2).

6. The method according to any of the preceding claims,wherein said measurement data are collected both at the startand at the end of the balancing process, and wherein a change incapacity of the battery system (1) over time is determined basedon stored data relating to a change in state of charge of each ofthe battery cell units (2) from the start to the end of the balancing process.

7. The method according to any of the preceding claims,wherein the impedance of each of the battery cell units (2) isdetermined based on measurement data collected in connectionwith switching in and out a known resistance (4) in parallel with each of the battery cell units (2).

8. The method according to claim 7, wherein a change inimpedance of the battery system (1) over time is determinedbased on stored data relating to the impedance of each of the battery cell units (2).

9. A monitoring unit (10) for monitoring a battery system (1)comprising a plurality of battery cell units (2) and a battery balancing module (8), the monitoring unit (10) comprising: 16 - measurement means (6, 7, 9) configured to collectmeasurement data from each of the battery cell units (2), - data storage means (11) for storing data in a database, - processing means (12) configured to, based on said measurement data, determine at least one performance related parameter for each of the battery cell units (2), and based on stored data relating to said at least oneperformance related parameter, determine a change in atleast one of capacity and impedance of the battery system(1) over time, characterized in that the monitoring unit (10) is configured to use measurement data collected during a balancing process of the battery system least related (1) to determine said at one performance parameter.

10. A computer program comprising computer program code forcausing a computer to implement a method according to any oneof the claims 1-8 when the computer program is executed in the computer.

11. A computer program product comprising a data storagemedium (44) which can be read by a computer and on which theprogram code of a computer program according to claim 10 is stored.

12. An electronic control unit (40) in a motor vehicle comprisingan execution means (41), a memory (42) connected to theexecution means (41) and a data storage medium (44) which is connected to the execution means (41) and on which the 17 computer program code of a computer program according toclaim 10 is stored.

13. A motor vehicle comprising an electronic control unit (40)according to claim 12.

14. A motor vehicle according to claim 13, wherein the motor vehicle is a truck, a bus or a passenger car.