Data-Driven Health Management of Electrical Vehicle Battery Systems

Project status


Start date: 10/01/14
End date: 06/29/18


Principal investigator:

Co-principal investigator:

About the research

The objectives of this research were to conduct theoretical and experimental investigations to develop a new battery health management paradigm based on a novel, self-cognizant dynamic system (SCDS) approach to predict and prevent failures of safety-critical battery systems (e.g., lithium plating and thermal runaway) for electric vehicles (EVs) and hybrid electric vehicles (HEVs) and develop an onboard diagnostics tool and alarm system for early awareness of these potential impending failures.

This research developed a technique that can adaptively recognize the dynamic characteristics of an operating battery system over time without relying on expensive, time-consuming battery tests for the prediction and prevention of safety-critical battery system failures. Battery failure prognostics employing the proposed SCDS-based health management paradigm can not only account for normal battery capacity fading over time but also identify abnormal safety-critical failures that usually happen in a relatively shorter time period.


Report: Data-Driven Health Management of Electrical Vehicle Battery Systems (867.06 kb pdf) August 2018

Related publication: Battery System Safety and Health Management for Electric Vehicles Aug 2015



  • Midwest Transportation Center
  • Wichita State University