Operation and efficiency improvements of a hybrid dual-active-bridge converter for energy storage systems

• المؤلف الرئيسي: Dong, Jiaxin
• مؤلفون آخرون: Josep Pou
• التنسيق: Thesis-Doctor of Philosophy
• اللغة: English
• منشور في: Nanyang Technological University 2024
• الموضوعات: Engineering; Energy storage systems; Dual-active-bridge converter; Efficiency optimization
• الوصول للمادة أونلاين: https://hdl.handle.net/10356/175463
• المؤسسة: Nanyang Technological University

This thesis focuses on optimizing the operation and efficiency of the dual-active-bridge (DAB) converter for high-power energy storage systems. A novel DAB-based topology and some modulation strategies are proposed and thoroughly investigated in Chapter 2. The hybrid neutral-point-clamped (NPC) modules in the DAB converter topology utilize Si IGBTs and SiC MOSFETs to reduce power losses while maintaining a cost-efficiency balance. A new modulation method is introduced in Chapter 3, applying different switching schemes for the SiC MOSFETs and the Si IGBTs in the hybrid NPC DAB converter. This design effectively utilizes the neutral current path and significantly reduces conduction and switching losses. Experimental comparisons demonstrate higher efficiency of the proposed hybrid duty ratio phase-shifted modulation (PSM) . To address modelling challenges, a novel data-driven modelling with experimental augmentation (D2EA) is proposed in Chapter 4, combining simulation and experimental data to improve accuracy. The D2EA achieves an efficiency modelling accuracy of 99.92% and is validated in 2-kW hardware experiments, reaching a peak efficiency of 98.45%. Chapter 5 presents an artificial intelligent- (AI-) based control method to optimize zero-voltage zero-current (ZVZCS) operation and improved efficiency performance. The hybrid NPC DAB converter demonstrates superior efficiency under various load conditions compared to conventional triple-phase-shifted (TPS) modulation. This research offers valuable insights into optimizing DAB converters, introducing novel strategies and control methods. The proposed D2EA approach and AI-based optimization show promise in advancing power converter modeling and efficiency, paving the way for future high-power energy storage system applications.