Robust and fast estimation of equivalent circuit model from noisy electrochemical impedance spectra

•Robust and fast ECM fitting algorithm has been developed and tested on noisy impedance spectra, with source code available at github.com/leehangyue/EISART.•Accurate Complex Nonlinear Least Square (CNLS) fit has been realized with DRT-based initialization of ECM parameters.•Robust DRT analysis and E...

Full description

Saved in:
Bibliographic Details
Published inElectrochimica acta Vol. 422; p. 140474
Main Authors Li, Hangyue, Lyu, Zewei, Han, Minfang
Format Journal Article
LanguageEnglish
Published Oxford Elsevier Ltd 01.08.2022
Elsevier BV
Subjects
Online AccessGet full text
ISSN0013-4686
1873-3859
DOI10.1016/j.electacta.2022.140474

Cover

More Information
Summary:•Robust and fast ECM fitting algorithm has been developed and tested on noisy impedance spectra, with source code available at github.com/leehangyue/EISART.•Accurate Complex Nonlinear Least Square (CNLS) fit has been realized with DRT-based initialization of ECM parameters.•Robust DRT analysis and ECM fitting has been enabled by data screening and weighing.•The estimation error of resistances in ECM from 1% to 26% were obtained for various noise levels and the factors affecting accuracy were revealed.•Time consumption fitting an ECM has been shortened to within 1 s on a typical personal laptop. The Equivalent Circuit Model (ECM) is a powerful technique to quantitatively analyze and compare Electrochemical Impedance Spectroscopy (EIS) data. In practice, noise is prevalent in EIS data, due to fuel cell system fluctuations, limited measuring time, and instruments, challenging the accuracy of ECM. There are algorithms that work well on noisy data, yet in many cases, either run time or robustness remains a problem. In this paper, we proposed a robust and fast algorithm that detects outliers, weighs the EIS data, and automatically fits an ECM within a second. For both experimentally measured and simulated noisy EIS data, the new algorithm reduced the impact of noise drastically. The algorithm is demonstrated in Python as part of the open-source software EISART at github.com/leehangyue/EISART. [Display omitted]
Bibliography:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:0013-4686
1873-3859
DOI:10.1016/j.electacta.2022.140474