Fault prognosis of Li-ion batteries in electric vehicles: Recent progress, challenges and prospects

Abstract

Lithium-ion (Li-ion) batteries have gained significant prominence in various applications, including electric vehicles (EVs), due to their long lifespan, high energy density , and power capabilities. However, these batteries are prone to faults that can adversely affect their performance and safety. Therefore, accurate and timely fault prognosis is crucial for identifying and mitigating these faults, ensuring the reliable and efficient operation of battery systems. This paper presents a comprehensive review of fault prognosis techniques for Li-ion batteries in EVs. It covers various types of faults, such as thermal runaway, short circuits, and voltage deviations, and discusses different prognostic approaches, including model-based methods, data-driven techniques, and hybrid models, along with their respective strengths and limitations. Furthermore, the paper explores state-of-the-art fault prognosis methods, including deep learning , transfer learning , multi-sensor data fusion , and prognostics-based health management. It reviews relevant case studies and application examples, providing valuable insights into their practical implementation and effectiveness in EV battery systems. The findings contribute to the development of effective strategies for enhancing the reliability, safety, and overall performance of Li-ion battery systems in EVs. • Focus on Li-ion battery fault prognosis, shifting from diagnosis-centric research. • Explore the correlation of faults with battery chemistry. • Address challenges, trends, and opportunities in fault prognosis, bridging gaps in the literature. • Offer comprehensive insight into battery faults and their internal mechanism. • Evaluate the existing fault prognosis methods, identify limitations, and propose solutions.