Seite - 172 - in Emerging Technologies for Electric and Hybrid Vehicles
Bild der Seite - 172 -
Text der Seite - 172 -
Energies 2017,10, 5
7LPH V
H[SHULPHQWDO GDWD
FRQYHQWLRQDO ILWWLQJ IXQFWLRQ
LPSURYHG ILWWLQJ IXQFWLRQ
(a) 7LPH V
H[SHULPHQWDO GDWD
FRQYHQWLRQDO ILWWLQJ IXQFWLRQ
LPSURYHG ILWWLQJ IXQFWLRQ
(b)
7LPH V
FRQYHQWLRQDO ILWWLQJ IXQFWLRQ
LPSURYHG ILWWLQJ IXQFWLRQ
(c)
Figure12.Voltagecurvesofonecycleof thedischargingpulse-rest test (62%–60%): (a) theoverview;
(b) aclose look; (c) thevoltageacross the long-termRCnetwork.
4.2.ModelVerification
In thispaper, theCCchargingtestandtheconsecutiveUDDStest,whichrespectivelyrepresent
two typical operating scenarios inHEV/EV applications, are conducted separately to verify the
effectiveness of themodel. For the charging condition, thebattery is charged from10%–90%SoC.
The typical chargingcurrent inpracticevaries fromC/8to2C[50], andaC/2ratecurrent isemployed
inthechargingtest. TheconsecutiveUDDStest starts from90%SoC to20%SoC,witha10-minrest
period inbetweentosimulateashortparkingtime. In therealapplication,aspecificsetofparameters
canbeselectedbythecharacteristicsof themeasuredloadcurrent. Forexample, if thevaluesof the
currentareapproximatelyconstantoveracertain time interval,parametersestimatedfromthedata in
thepulse-chargingperiodsareemployed.Ontheotherhand,parametersestimatedfromthedata in
therestperiodsareemployedwhenthe loadcurrentshowsthecharacteristicsofhighdynamicsovera
certain timeinterval.
Firstly, for theCCchargingscenario, threemodeloutputsandmeasuredbattery terminalvoltage
curvesareplotted inFigure13,andthecorrespondingRMSEsaregiven inTable4. It canbeobserved
that during thewhole charging process, themodel with parameters estimated from the data in
pulse-charging periods outputs a voltage curvematching themeasured curve better because of
consideringthecontinuousexternalelectricdrivingforces.However,parametersestimatedfromthe
data in therestperiodsresult inrelatively largererrors, especially in thehighSoC region. Inaddition,
duringmostpart of the chargingperiod, themodelwithparametersused in thedynamicdriving
scenarios outputs a voltage higher than the experimental voltage. Comparing the corresponding
curves inFigures8band11, it canbededucedthat thehigherestimatedvoltage ismainlycausedby
the largervalueofestimatedRlong, especiallyduringthemiddlerangeof theSoC region.
Table4.RMSEofmodelvoltageestimationunder theCCchargingtest.
ModelingMethods DynamicCondition Rest-Period Pulse-Period
RMSE(mV) 18.41 19.76 5.448
172
Emerging Technologies for Electric and Hybrid Vehicles
- Titel
- Emerging Technologies for Electric and Hybrid Vehicles
- Herausgeber
- MDPI
- Ort
- Basel
- Datum
- 2017
- Sprache
- englisch
- Lizenz
- CC BY-NC-ND 4.0
- ISBN
- 978-3-03897-191-7
- Abmessungen
- 17.0 x 24.4 cm
- Seiten
- 376
- Schlagwörter
- electric vehicle, plug-in hybrid electric vehicle (PHEV), energy sources, energy management strategy, energy-storage system, charging technologies, control algorithms, battery, operating scenario, wireless power transfer (WPT)
- Kategorie
- Technik