Statistical analysis for understanding and predicting battery degradations in real-life electric vehicle use

Anthony Barré (1, 2) , Frédéric Suard (1) , Mathias Gérard (2) , Maxime Montaru (2) , Delphine Riu (3)

1 LADIS (CEA, LIST) - Laboratoire d'analyse des données et d'intelligence des systèmes (CEA, LIST)
2 LITEN - Laboratoire d'Innovation pour les Technologies des Energies Nouvelles et les nanomatériaux
3 G2ELab - Laboratoire de Génie Electrique de Grenoble

This paper describes the statistical analysis of recorded data parameters of electrical battery ageing during electric vehicle use. These data permit traditional battery ageing investigation based on the evolution of the capacity fade and resistance raise. The measured variables are examined in order to explain the correlation between battery ageing and operating conditions during experiments. Such study enables us to identify the main ageing factors. Then, detailed statistical dependency explorations present the responsible factors on battery ageing phenomena. Predictive battery ageing models are built from this approach. Thereby results demonstrate and quantify a relationship between variables and battery ageing global observations, and also allow accurate battery ageing diagnosis through predictive models.

Mots clés: Li-ion battery; Battery; degradation; Ageing; Prediction; Statistics; Battery degradation; Electric Vehicle

Domaines: Energie électrique; Modélisation et simulation

Anthony Barré, Frédéric Suard, Mathias Gérard, Maxime Montaru, Delphine Riu. Statistical analysis for understanding and predicting battery degradations in real-life electric vehicle use. Journal of Power Sources, 2014, 245, pp.846-856. ⟨10.1016/j.jpowsour.2013.07.052⟩. ⟨hal-01071585⟩