Project Description

Advances in electrochemical energy storage are critical toward enabling vehicle electrification and renewable energy integration into the electric grid. Recent years have witnessed an urgent need to accelerate innovation toward realizing prolonged and safe utilization of high energy and power density batteries (e.g., lithium-ion batteries for electric vehicles and grid storage). Despite significant advances in high-throughput experimental and model-based characterization, accelerated prognostics of life and safe operation of batteries remains a critical challenge. This research will entail machine learning enabled prognostics with integrated Li-ion battery cycling data and simulated degradation from physics-based models based on operational variability.
The candidate will collaborate with an interdisciplinary team from Mechanical Engineering (Partha P. Mukherjee), Chemical Engineering (Brett Savoie), and Mathematics/Statistics (Guang Lin) with expertise in energy storage, data sciences and machine learning.

Start Date

June 01, 2021

Postdoc Qualifications

Degree in engineering, or computational and data science or applied math/stat
Knowledge with data-driven and machine learning approaches
Motivated and team-player to work in an interdisciplinary research setting 

Co-Advisors

Partha P. Mukherjee, Associate Professor, School of Mechanical Engineering
pmukherjee@purdue.edu
https://engineering.purdue.edu/ETSL/

Brett Savoie, Assistant Professor, Davidson School of Chemical Engineering
bsavoie@purdue.edu
https://engineering.purdue.edu/savoiegroup/

Guang Lin, Associate Professor, Departments of Mathematics/Mechanical Engineering, Statistics(Courtesy)
guanglin@purdue.edu
http://www.math.purdue.edu/~lin491/ 

References

A. Mistry, K. Smith, and P. P. Mukherjee, “Stochasticity at Scales Leads to Lithium Intercalation Cascade,” ACS Applied Materials and Interfaces, 12, 16359 (2020).

A. Mistry and P. P. Mukherjee, “Deconstructing electrode pore network to learn transport distortion,” Physics of Fluids, 31, 122005 (2019).

N. C. Iovanac, B. M. Savoie. Simpler is Better: How Linear Prediction Tasks Improve Transfer Learning in Chemical Autoencoders J. Phys. Chem. A (2020), vol. 124 (18), 3679–3685.

S. Zhang, and G. Lin, “Robust data-driven discovery of governing physical laws with error bars,” Proc. Royal Soc. A. 474, 20180305 (2018).