Numerical Modeling of Lithium Ion Battery Capacity Fading on Electric Vehicles

Abstract

Electric vehicles are recognized as clean alternative to conventional vehicles and promoted globally for large fleet deployment. As predicted, the EV global fleet number will reach 2.6 million by 2016 and 230 million by 2030. Current EVs are all powered by lithium ion batteries due to its higher energy density and power density. There are two types of EVs on current market: Plug-in Hybrid EV, represented by Chevrolet Volt, and battery EV, represented by Nissan Leaf. The lithium ion batteries used on both Nissan Leaf and Chevrolet Volt are with the same chemistry: graphite anode and lithium manganese oxides (LMO) cathode. However, current EV lithium ion batteries have severe capacity fading issue which limits the market penetrations of EVs. For example, the Nissan Leaf battery pack, with 24 kWh capacity, can power the vehicle for 84 miles/charge and the battery performance is warranted for 5 years or 60,000 miles only. In this research, we report our numerical modeling approach for predictive modeling of battery capacity fading on electric vehicle operations for both cycling capacity loss and calendar capacity loss. The capacity fading of a typical 24 kWh LMO-Graphite lithium ion battery during actual EV operations are mathematically modeled under state-level average driving conditions in the United States, considering the EV travel demand, driving range, hourly temperature, and driving pattern. The calculated annual cycling capacity losses are between 0.5% and 1.2% in first year, and slightly going up in following years. The calculated annual calendar capacity losses are between 4% and 8.9% in first year, and falling down quickly in following years. The results are validated by actually reported Nissan Leaf capacity fading data in several representative states of U.S.