Dynamic Control and DC Fault Ride-Through of Double-Stage Solid-State Transformers

• Applicant: Professor Dr. Marcelo Lobo Heldwein
• Subject Area: Electrical Energy Systems, Power Management, Power Electronics, Electrical Machines and Drives
• Term: since 2025
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 560598562

The solid-state transformer (SST) is a promising way to replace traditional line-frequency transformers and relevant power electronic components, particularly in applications such as hydrogen generation systems and ultrafast charging stations. It not only offers the basic functions of traditional line-frequency transformers in a more compact form but also enhances the autonomy of nodes in distribution systems due to its smart functions and capabilities. The SST normally consists of a rectifier stage and an isolated DC stage, converting medium-voltage AC to low-voltage DC. However, due to the different bandwidths and response speeds of the two stages, overvoltage and undervoltage conditions may occur, potentially leading to overmodulation of the rectifier stage and overcurrent in the DC stage. More seriously, this condition could endanger the reliable operation of the entire system. To address this issue, this proposal presents an integrated model predictive control framework, which significantly enhances the dynamic performance of the SST, providing a nearly seamless transient switching capability, even under fault conditions. Moreover, an advanced observer based on super-twisting method with quantitative sensitivity analysis will be developed in the project, improving the robustness of the control system. Additionally, this proposal introduces a comprehensive method that enables SSTs to ride through DC fault conditions safely, quickly, and economically, and to resume regular operation in minimal time once the fault is cleared. During fault events, this method simultaneously ensures a criterion faulty current, accurately controlled transformer current impact, balanced modular input voltage, and elimination of DC bias. Meanwhile, the same control framework is maintained for both regular and fault conditions, avoiding the need for switching algorithms. This method provides greater flexibility for SST cooling, high-frequency transformer design, and component selection while maintaining a simple control system. The project is divided into four work packages and will be studied and validated through theoretical analysis, simulations, and experiments. The proposed methods will be comprehensively tested under varying levels of parameter errors and fault conditions. Finally, the method will be compared and evaluated comprehensively against existing approaches.

• DFG Programme: Research Grants
• Co-Investigators: Professor Dr.-Ing. Ralph Kennel; Dr. Gean Jacques Maia de Sousa