This thesis presents the design, simulation and experimental validation of an Energy Management System (EMS) for a Hybrid Energy Storage System (HESS) composed of lithium ion batteries and Supercapacitors (SCs) in electric vehicles. The aim of the EMS is to split the power demand considering the weaknesses and strengths or the power sources. The HESS requires an EMS to determine power missions for the battery and SC in real time, where the SC is commanded to assist the battery during high power demand and recover the energy generated during braking. Frequency sharing techniques have been proposed by researchers to achieve this objective, including the Discrete Wavelet Transform (DWT) and conventional filtration methods (low and high pass filters). However, filtration approaches can introduce delay (milliseconds to tens of seconds) in the frequency components which undermines the hybridisation advantages. Hence, the selection of the filtration technique and filter design are crucial to the system's performance. Researchers have proposed power demand prediction methodologies to deal with time delay, however, the advantages and drawbacks of using such methods have not been investigated thoroughly, particularly whether time delay compensation and its inherent prediction error improves the system performance, efficiency, and timely SC contribution during the motoring and braking stages. This work presents a fresh perspective to this research field by introducing a novel approach that deals with delay without complicated prediction algorithms and improves the SC contribution during the motoring and braking stages while reducing energy losses in the system. The proposed EMS allows the SC to provide timely assistance during motoring and to recover the braking energy generated. A charging strategy controls energy circulation between the battery and SC to keep the SC charge availability during the whole battery discharge cycle. The performance and efficiency of the HESS is improved when compared to the traditional use of conventional filtration techniques and the DWT. Results show that the proposed EMS method improves the energy efficiency of the HESS. For the US06 driving cycle, the energy efficiency is 91.6%. This is superior to the efficiency obtained with an EMS based on a high pass filter (41.3%), an EMS based on DWT high frequency component (30.3%) and an EMS based on the predicted DWT high frequency component (41%).