TY - JOUR
T1 - Model-Based Battery Thermal Fault Diagnostics
T2 - Algorithms, Analysis, and Experiments
AU - Dey, Satadru
AU - Perez, Hector E.
AU - Moura, Scott J.
N1 - Funding Information:
Manuscript received July 21, 2017; revised October 30, 2017; accepted November 12, 2017. Date of publication December 6, 2017; date of current version February 8, 2019. Manuscript received in final form November 17, 2017. This work was supported by the National Science Foundation under Grant 1408107. Recommended by Associate Editor A. G. Stefanopoulou. (Corresponding author: Satadru Dey.) S. Dey is with the Department of Electrical Engineering, University of Colorado, Denver, CO 80204 USA (e-mail: satadru.dey@ucdenver.edu).
Publisher Copyright:
© 1993-2012 IEEE.
PY - 2019/3
Y1 - 2019/3
N2 - Safety and reliability remain critical issues for lithium-ion (Li-ion) batteries. Out of many possible degradation modes, thermal faults constitute a significant part of critical causes that lead to battery degradation and failure. Therefore, it is extremely important to diagnose these thermal faults in real time to ensure battery safety. Motivated by this fact, we propose a partial differential equation (PDE) model-based real-time scheme in this paper for diagnosing thermal faults in Li-ion batteries. The objective of the diagnostic scheme is to detect and estimate the size of the thermal fault. We utilize a distributed parameter 1-D thermal model for cylindrical battery cells in conjunction with PDE observer-based techniques to design the scheme. Furthermore, we apply threshold-based technique to ensure robustness against modeling and measurement uncertainties. The effectiveness of the scheme is illustrated by: 1) analytical convergence verification of the PDE observers under heathy and faulty conditions utilizing Lyapunov stability theory; 2) extensive simulation case studies; 3) robustness analysis against model parametric uncertainties; and 4) experimental studies on a commercial Li-ion battery cell.
AB - Safety and reliability remain critical issues for lithium-ion (Li-ion) batteries. Out of many possible degradation modes, thermal faults constitute a significant part of critical causes that lead to battery degradation and failure. Therefore, it is extremely important to diagnose these thermal faults in real time to ensure battery safety. Motivated by this fact, we propose a partial differential equation (PDE) model-based real-time scheme in this paper for diagnosing thermal faults in Li-ion batteries. The objective of the diagnostic scheme is to detect and estimate the size of the thermal fault. We utilize a distributed parameter 1-D thermal model for cylindrical battery cells in conjunction with PDE observer-based techniques to design the scheme. Furthermore, we apply threshold-based technique to ensure robustness against modeling and measurement uncertainties. The effectiveness of the scheme is illustrated by: 1) analytical convergence verification of the PDE observers under heathy and faulty conditions utilizing Lyapunov stability theory; 2) extensive simulation case studies; 3) robustness analysis against model parametric uncertainties; and 4) experimental studies on a commercial Li-ion battery cell.
UR - http://www.scopus.com/inward/record.url?scp=85038376510&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85038376510&partnerID=8YFLogxK
U2 - 10.1109/TCST.2017.2776218
DO - 10.1109/TCST.2017.2776218
M3 - Article
AN - SCOPUS:85038376510
VL - 27
SP - 576
EP - 587
JO - IEEE Transactions on Control Systems Technology
JF - IEEE Transactions on Control Systems Technology
SN - 1063-6536
IS - 2
M1 - 8166763
ER -