Battery Management Systems. Accurate State-of-Charge Indication for Battery-Powered Applications

Transcription

1

2 Battery Management Systems Accurate State-of-Charge Indication for Battery-Powered Applications

3 Philips Research VOLUME 9 Editor-in-Chief Dr. Frank Toolenaar Philips Research Laboratories, Eindhoven, The Netherlands SCOPE TO THE PHILIPS RESEARCH BOOK SERIES As one of the largest private sector research establishments in the world, Philips Research is shaping the future with technology inventions that meet peoples needs and desires in the digital age. While the ultimate user benefits of these inventions end up on the high-streeet shelves, the often pioneering scientific and technological basis usually remains less visisble. This Philips Research Book Series has been set up as a way for Philips researchers to contribute to the scientific community by publishing their comprehensive results and theories in book form. Dr. Rick Harwig

4 Battery Management Systems Accurate State-of-Charge Indication for Battery-Powered Applications by Valer Pop Holst Centre/IMEC-NL Eindhoven, The Netherlands Henk Jan Bergveld NXP Semiconductors Eindhoven, The Netherlands Dmitry Danilov Eurandom, Eindhoven The Netherlands Paul P.L. Regtien University of Twente Enschede, The Netherlands and Peter H.L. Notten Philips Research Laboratories Eindhoven Eindhoven University of Technology Eindhoven, The Netherlands

5 Valer Pop Holst Centre/IMEC-NL Eindhoven, The Netherlands Henk Jan Bergveld NXP Semiconductors Eindhoven, The Netherlands Dmitry Danilov Eurandom Eindhoven, The Netherlands Paul P.L. Regtien University of Twente Enschede, The Netherlands Peter H.L. Notten Philips Research Laboratories Eindhoven Eindhoven University of Technology Eindhoven, The Netherlands ISBN e-isbn Library of Congress Control Number: Springer Science+Business Media B.V. No part of this work may be reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, microfilming, recording or otherwise, without written permission from the Publisher, with the exception of any material supplied specifically for the purpose of being entered and executed on a computer system, for exclusive use by the purchaser of the work. Printed on acid-free paper springer.com

6 To Raluca, Peggy, Olga, Elly and Pascalle

7 Table of contents List of abbreviations List of symbols xi xiii 1. Introduction Battery Management Systems State-of-Charge definition Goal and motivation of the research described in this book Scope of this book References 7 2. State-of-the-Art of battery State-of-Charge determination Introduction Battery technology and applications General operational mechanism of batteries Battery types and characteristics Summary History of State-of-Charge indication A general State-of-Charge system Possible State-of-Charge indication methods Direct measurement Book keeping systems Adaptive systems Summary Commercial State-of-Charge indication systems Conclusions References A State-of-Charge indication algorithm An introduction to the algorithm Battery measurements and modelling for the State-of-Charge indication algorithm EMF measurement and modelling Overpotential measurement and modelling States of the State-of-Charge algorithm Main issues of the algorithm EMF measurement, modelling and implementation Overpotential measurement, modelling and implementation Adaptive systems General remarks on the accuracy of SoC indication systems Conclusions References Methods for measuring and modelling a battery s Electro-Motive Force EMF measurement Voltage prediction 69

8 viii Table of contents Equilibrium detection Existing voltage-relaxation models used for voltage prediction A new voltage-relaxation model Implementation aspects of the voltage-relaxation model Comparison of results obtained with the different voltage-relaxation models Summary 4.3 Hysteresis Electro-Motive Force modelling Conclusions References Methods for measuring and modelling a battery s overpotential Overpotential measurements Overpotential measurements involving partial charge/discharge steps Overpotential measurements involving full (dis)charge steps Overpotential modelling and simulation Overpotential modelling Simulation results Conclusions References Battery aging process General aspects of battery aging Li-ion battery aging Q max measurements EMF measurements as a function of battery aging The voltage-relaxation model as a function of battery aging EMF GITT measurement results obtained for aged batteries The charge/discharge Electro-Motive Force difference as a function of battery aging EMF modelling as a function of battery aging Overpotential dependence on battery aging Overpotential measurements as a function of aging Adaptive systems Electro-Motive Force adaptive system Overpotential adaptive system Conclusions References Measurement results obtained with new SoC algorithms using fresh batteries Introduction Implementation aspects of the algorithm 146

9 Table of contents ix A new SoC algorithm Implementation aspects of the SoC algorithm Results obtained with the algorithm using fresh batteries Uncertainty analysis Uncertainty in the real-time SoC evaluation system The SoC uncertainty The remaining run-time uncertainty Improvements in the new SoC algorithm A new State-of-Charge-Electro-Motive Force relationship A new State-of-Charge-left model Determination of the parameters of the new models Test results Uncertainty analysis Comparison with Texas Instruments bq26500 SoC indication IC The bq26500 SoC indicator Comparison of the two SoC indicators Conclusions References Universal State-of-Charge indication for battery-powered applications Introduction Implementation aspects of the overpotential adaptive system SoC=f(EMF) and SoC l adaptive system Results obtained with the adaptive SoC system using aged batteries Uncertainty analysis Results obtained with other Li-based battery EMF and SoC l modelling results obtained for the Li-based battery Experimental results Practical implementation aspects of the SoC algorithm Hardware design of the evaluation board Software design of the evaluation board Measurement results Boostcharging Conclusions References General conclusions 221

10 List of abbreviations ADC Analog-to-Digital Converter ANN Artificial Neural Networks AR d Actual Rate discharge current ACU Adaptive Control Unit b Boostcharge BMS Battery Management System BKM Book-Keeping Module cc Coulomb counting CC Constant-Current CCCV Constant-Current-Constant-Voltage CCCVCCCV (CCCV) 2 CD Compact Disc C n Cycle number ch Charge CCA Charging Current Accumulator CAC Compensated Available Charge d Discharge DAC Digital-to-Analog Converter DCA Discharging Current Accumulator DAQ Data Acquisition interface card Dig. I/O Digital Input/Output pins EIS Electrochemical Impedance Spectroscopy EMF dt EMF detection method EMF f EMF fitting method EKF Extended Kalman Filters EEPROM Electrically Erasable Read-Only Memory EMF m Measured EMF data points f sd Self-discharge rate estimation rate GUI Graphical User Interface GITT Galvanostatic Intermittent Titration Technique HDQ High-speed single-wire interface HEV Hybrid Electrical Vehicles ISP In-System Programming IF In-Functional IC Integrated Circuit I 2 C Inter-Integrated-Circuit Bus ICA Integrated Current Accumulator JTAG Joint Test Action Group KF Kalman Filters LED Light-Emitting Diode LCD Liquid-Crystal Display Li-ion Lithium-ion LA Lead-Acid Li-ion POL Lithium-ion Polymer Li-SO 2 Lithium Sulphur Dioxide Li Lithium LCR Learning Count Register

11 xii List of abbreviations LMD NiCd NiMH NI NAC NTC N OLS PGA PM ph PC RTOS ROM RAM SoH SBS sd SEI SCB-68 SPI UART V pm VDQ η m Last Measured Discharge Nickel Cadmium Nickel Metal Hydride National Instruments Nominal Available Charge Negative-Temperature-Coefficient Number of ADC bits Ordinary Least Squares Programmable Gain Amplifier Power Module Phase transition Personal Computer Real-Time Operating System Read-Only Memory Random-Access Memory State-of-Health Smart Battery System Self-discharge Solid Electrolyte Interface National Instruments connector board Serial Peripheral Interface Universal Asynchronous Receiver Transmitter Voltage prediction model Valid Discharge flag Overpotential model

12 List of symbols Symbol Meaning Value Unit a aged A Parameter for the SoC-EMF [1] a 10 Parameter for the SoC-EMF [1] a 11 Parameter for the SoC-EMF [1] a 12 Parameter for the SoC-EMF [1] a 20 Parameter for the SoC-EMF [1] a 21 Parameter for the SoC-EMF [1] a 22 Parameter for the SoC-EMF [1] c o Parameter for overpotential [Ω A 1 ] modelling c 1 Parameter for overpotential [V] modelling c 2 Parameter for overpotential 1 [ s ] modelling c 3 Parameter for overpotential [J] modelling c 4 Parameter for overpotential 1 [A 1 ] modelling c 5 Parameter for overpotential [A] modelling C d Mean discharge C-rate current DoD Depth-of-Discharge [%] DoC Depth-of-Charge [%] EMF Electro-Motive Force [V] o E q Amount of the energy that [J] cannot be obtained from the battery I E q Non-linear part of the amount of the energy that cannot be [J] obtained from the battery + E eq Equilibrium potential of the [V] positive electrode E eq Equilibrium potential of the [V] negative electrode

13 xiv List of symbols Symbol Meaning Value Unit + E 0 Standard redox potential of the [V] positive electrode E Standard redox potential of [V] 0 the negative electrode EDV 1 End-of Discharge Voltage level [V] EMF Estimated EMF model [V] es m EMF a5.4% EMF obtained for the 5.4% capacity loss battery (a=aged) EMF a25.4% EMF obtained for the 25.4% capacity loss battery (a=aged) EMF 1 First EMF point determined for [V] the EMF adaptation EMF ad EMF retrieved by means of the [V] adaptation method E 0 Parameter for the SoC-EMF [V] EMF p Predicted EMF voltage [V] EMF GITT EMF measurement by means of the GITT EMF f Fitted EMF [V] F Faraday constant [C mol 1 ] f Frequency [Hz] f Fresh I Constant maximum current max s (s=standard) [A] min I s Predefined minimum current (s=standard) [A] I Maximum boostcharging [A] max b current I Current that flows into(out) [A] of the battery I M Uncertainties from Maccor Neglected [A] current measurements I s Standby current [A] I b Backlight-on state current [A] I d Measured discharge current [A] I lim Limit current [A] M Slope of second asymptote of prior-art voltage-prediction method n 0 Parameter related to the magnitude [1] of the diffusion overpotential n 1 Parameter related to the magnitude [T 1 ] of the diffusion overpotential

14 List of symbols xv Symbol Meaning Value Unit OCV Open-Circuit Voltage [V] OCV f Battery OCV for a fresh battery [V] OCV a Battery OCV for an aged battery [V] par(t ref ) Value of one of the EMF-SoC [1] model parameters at temperature T ref p 11 Parameter for the SoC-EMF [1] p 12 Parameter for the SoC-EMF [1] p 21 Parameter for the SoC-EMF [1] p 22 Parameter for the SoC-EMF [1] q 11 Parameter for the SoC-EMF [1] q 12 Parameter for the SoC-EMF [1] q 21 Parameter for the SoC-EMF [1] q 22 Parameter for the SoC-EMF [1] Q in Charge present in the battery [Ah] at the time t Q d2 Discharge capacity for battery [Ah] number 2 Q d1 Discharge capacity for battery [Ah] number 1 1 Q d1 Q d1 value after the first cycle [Ah] 220 Q d1 Q d1 value after 220 cycles [Ah] 1 Q d 2 Q d2 value after the first cycle [Ah] 2000 Q d 2 Q d2 value after 2000 cycles [Ah] Q max Battery maximum capacity [Ah] Q max1 Maximum capacity of battery [Ah] Q max2 Maximum capacity of battery [Ah] Q loss Capacity loss [%] Q d Discharge battery capacity [Ah] Q dd Decrease in Q d [%] + Q max Maximum capacity of the positive 1.00 [1] electrode Q max Maximum capacity of the negative [1] electrode

15 xvi List of symbols Symbol Meaning Value Unit Q max Amount of the electrochemically [1] + active Li ions inside the battery Q maxr Reference maximum capacity value [Ah] Q Amount of the Li + ions that will [1] 0 remain in the negative electrode after discharging Q ch Amount of charge flowing into [Ah] the battery during the charge state Q d Discharge capacity [Ah] R Gas constant J(mol K) 1 ) R Resistance [Ω] 0 R d Linear contribution of the [Ω] diffusion resistance I R d Non-linear contribution of the [Ω] diffusion resistance R S Sense resistor [Ω] R Ωk Linear contribution of the ohmic [Ω] and kinetic resistance R Ik Non-linear contribution of the [Ω] ohmic and kinetic resistance R HF High-frequency resistance [Ω] s x Sign of parameter x SoC State-of-Charge [%] SoC(EMF) SoC calculated based on the [%] EMF voltage SoC(V p ) SoC calculated based on the [%] predicted voltage SoC l SoC-left [%] es lm SoC Estimated SoC l model [%] SoC lm Measured SoC l [%] SoC lf Fitted SoC l [%] SoC e SoC error [%] SoC si Initial SoC in the standby state [%] SoC sf Final SoC in the standby state [%] SoC st SoC indicated at the start [%] SoC end SoC indicated at the end [%] SoC in SoC in the initial state [%] SoCs SoC in the standby state [%] SoC t SoC in the transitional state [%] SoC ch SoC in the charge state [%] SoC d SoC in the discharge state [%] SoC b SoC in the backlight-on state [%]