Enhanced Bus Voltage Stability through Digital Twin-Enabled Adaptive Controller Tuning
In this paper, a digital twin approach for adaptive controller tuning is presented. For an anticipated load profile, step transitions are fed into a simulation-based optimizer which utilizes a state-space averaged model to determine optimal control parameters for the future operating point after the next upcoming load step. This process aims to minimize voltage overshoot, settling time, and oscillation. Control parameters are adjusted preemptively before each load step by the simulation-based optimizer and subsequently validated through a comprehensive switching model simulation. This guarantees good dynamic performance throughout the load profile. Experimental results are provided for a simple example involving a two-loop PI controlled bidirectional DC/DC converter connected to a pulse load. Compared to the static PI controller, the digital twin tuning approach reduced voltage overshoot by ~50% and settling time by ~50% for the single converter. Simulation results are presented for a multi-converter system with a complex load profile, showing significant improvement, with up to 96% improvement in voltage settling time and up to 95% improvement in voltage overshoot.
Authors
- M. Bélanger
- Kerry Sado (ORCID: https://orcid.org/0000-0002-5565-4795)
- Enrico Santi (ORCID: https://orcid.org/0000-0001-6135-0096)
- Andy Wong
Institutions
- University of South Carolina (US)
Publication Details
- Published
- 2025-03-16
- DOI
- https://doi.org/10.1109/apec48143.2025.10977085
- Citations
- 1
- Primary Topic
- Real-time simulation and control systems
- Type
- article
- Field-Weighted Citation Impact
- 1.21
Funders
- Office of Naval Research