The lithium-ion battery in electric vehicle may decay quickly due to vibration and aging. One purpose of this research is to diagnose the degradation mechanism under such conditions. The other is to realize the state of health ( SOH) prediction. The specific methods are given as follows. First, the attenuation results of the battery affected by vibration are analyzed. Secondly, the degradation mode is identified. Then, the incremental capacity-differential voltage ( IC-DV) curve is used to quantify the degradation mode. The results of group Z are illustrated as follows. The loss of active material is 36. 94% , the loss of lithium ions is 35. 12% , and the conductivity loss is 1. 9% . Finally, the quantified results are taken as the input to formulate the GA-Elman model to realize SOH prediction. The errors are within 5% , which can meet the requirements of battery management system (BMS). This research provides a basis for the diagnosis and SOH prediction of lithium-ion batteries under vibration and aging conditions. It could help BMS formulate relevant strategies to extend battery life.