A wavelet-transform-based strategy is proposed for the power management of hybrid electric vehicles (HEV) with multiple on-board energy sources and energy storage systems including a battery, a fuel cell, and an ultra-capacitor. The proposed wavelet-transform algorithm is capable of identifying the high-frequency transient and real time power demand of the HEV, and allocating power components with different frequency contents to corresponding sources to achieve an optimal power management control algorithm. By using the wavelet decomposition algorithm, a proper combination can be achieved with a properly sized Ultra-capacitor dealing with the chaotic high-frequency components of the total power demand, while the fuel cell and battery deal with the low and medium frequency power demand. Thus the system efficiency and life expectancy can be greatly extended. Simulation and experimental results validated the effectiveness of wavelet-transform-based power management algorithm. (C) 2008 Elsevier B.V. All rights reserved.