Switched Affine System-Based Current Sharing Control in Interleaved VRM: A Lyapunov-LMI Approach for Compute Chip Power Delivery
To meet the high current supply demands of compute chip, the interleaved parallel buck converter has become the core design of the voltage regulation module (VRM) due to its high current supply capability, low output ripple, and high system efficiency. However, its performance heavily depends on the current sharing capability under both steady and transient conditions. Existing current sharing control strategies face challenges related to insufficient performance and stability, especially in power supply scenarios with the current load characteristics of compute chip. Therefore, this paper proposes a current sharing control strategy based on a switched affine system model. This strategy achieves dynamic inter-phase compensation in a frequency-adaptive manner and realizes the decoupling regulation of voltage and current, resulting in a marked improvement in current sharing performance. Leveraging Lyapunov stability theory, the controller is designed by solving a linear matrix inequality (LMI), and the system’s global stability is formally proven. The effectiveness of the proposed method is verified through both simulation and experiments. The experimental comparison results demonstrate that the proposed control strategy significantly enhances dynamic response, current sharing accuracy, and system stability, providing new insights for optimizing VRM performance under current load conditions.
Authors
- Guidong Zhang (ORCID: https://orcid.org/0000-0002-2371-2890)
- Wei Hu (ORCID: https://orcid.org/0000-0003-2816-0244)
- Xiaobo Meng (ORCID: https://orcid.org/0000-0002-9453-7671)
- Xiaokai Su
- Y. Zeng
Institutions
- Guangdong University of Technology (CN)
- Guangzhou University (CN)
- Guangzhou Experimental Station (CN)
Publication Details
- Journal
- IEEE Journal of Emerging and Selected Topics in Power Electronics
- Published
- 2025-08-04
- DOI
- https://doi.org/10.1109/jestpe.2025.3595202
- Primary Topic
- Advanced Memory and Neural Computing
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- Natural Science Foundation of Guangdong Province for Distinguished Young Scholars