There is a problem of the life attenuation of lithium-ion batteries caused by high-frequency demand power components during electric vehicle driving. Thus a complete ensemble empirical mode decomposition-permutation entropy (CEEMD-PE) energy management strategy is proposed. The electric vehicle power demand is decomposed into a finite number of intrinsic mode functions (IMFs), and each component is reconstructed into low- and high frequency components according to the data complexity of each IMF measured by PE. Finally, the high-frequency components containing transient power and fast changes are assigned to the ultracapacitor, while the low-frequency components are assigned to the battery. The experimental results show that, compared with the Haar wavelet strategy, the root mean square (RMS) value of the lithium-ion battery current in the medium-speed and high-speed regions is reduced by 5.91% and 4.17%, respectively, and the current peak value is reduced by 14.70% and 5.77%, respectively. The proposed strategy helps to suppress the high-frequency power demand endured by lithium-ion batteries and reduce the impact of large currents on lithium-ion batteries.