Experimental study of fractional-order models for lithium-ion battery and ultra-capacitor: Modeling, system identification, and validation. (15th November 2020)

• Record Type: Journal Article
• Authors: Wang, Yujie; Li, Mince; Chen, Zonghai
• Is Part Of: Applied energy. Volume 278(2020)
• Publication Date: 2020-11-15
• Subjects: Fractional-order modeling -- Electrochemical impedance spectroscopy -- Frequency domain analysis -- System Identification -- Global optimization; Power (Mechanics) -- Periodicals; Energy conservation -- Periodicals; Energy conversion -- Periodicals; 621.042
• DOI: 10.1016/j.apenergy.2020.115736
• Languages: English
• ISSNs: 0306-2619

Abstract

Highlights: Experiments of electrochemical impedance spectroscopy are performed. Effects of temperature and SOC on impedance spectra are analyzed. Fractional-order models for lithium-ion batteries and ultra-capacitors are built. The global optimization is proposed for model parameter identification. The model accuracy is proved by experiments at different temperatures. Abstract: In recent years, with the promotion of new energy vehicles and micro-grids, energy storage technology has achieved considerable development. The hybridization of the lithium-ion batteries with ultra-capacitors can effectively improve the power capability of the energy storage system, which is suitable for the vehicle and micro-grid applications. High precision modeling is the foundation of the state estimation and energy management of the energy storage system. The fractional-order model provides applicable computational complexity and accuracy. This paper presents the experimental study of the fractional-order models used for lithium-ion batteries and ultra-capacitors. In this work, the experimental results of the electrochemical impedance spectroscopy of both lithium-ion batteries and ultra-capacitors are systematically analyzed. Then the temperature compensating fractional-order models of the batteries and ultra-capacitors are developed. Moreover, the particle swarm optimization based global optimization is proposed for on-line parameter identification with strict constraints. Therefore the parameters of the fractional-order models can be extracted from daily vehicle operating conditions. Finally, the accuracy of the presented model is proved by dynamic cycle conditions at different temperatures. The results indicate that the proposed method can well approximate the voltage of both the lithium-ion batteries and the ultra-capacitors with mean relative errors less than 4% and 3%, respectively.