Multi-Objective Optimization-Based Real-Time Control Strategy for Battery/Ultracapacitor Hybrid Energy Management Systems

Abstract

Battery and ultracapacitor (UC) have complementary advantages, which meet the requirements of energy storage systems (ESSs) for plug-in electric vehicles (PEVs). In this paper, a novel control strategy is proposed to manage the power distribution between the battery and UC for a hybrid energy management system (HEMS) in PEVs. This control strategy aims at realizing less power loss, longer battery lifecycle, as well as UC's stable terminal voltage and ability of quick charge/discharge. Based on these three optimization targets, we define three sets of loss functions and formulate a multi-objective optimization (MOO) problem. In particular, two different methods, weighted method and no-preference method, are implemented to transform the MOO problem into a uni-objective convex optimization problem. The final problem is solved using the Karush-Kuhn-Tucker conditions. Simulation is conducted to verify the proposed control strategy, while a battery-only ESS and a HEMS utilizing rule-based control strategy are implemented as a comparative study. A scaled-down laboratory prototype is built to validate the theoretical analysis and simulation results. The results indicate that the proposed control strategy brings the benefits of minimized battery current magnitudes and ripples, enhanced system efficiency, stabilized dc-link voltage, and improved dynamic response. Moreover, this strategy exhibits fast computation speed and requires no pre-information of future load demand. Therefore, it can be easily deployed in real-time.