Tsinghua University Invited Seminar: Digital Energy Processing and Computing—Theory and Practice

Overview

About this event

On October 30, 2026, the Center hosted an international expert seminar at Yonsei University featuring Professor Song Ci of Tsinghua University. Titled “Digital Energy Processing and Computing: Theory and Practice,” the seminar introduced digital energy processing concepts and battery energy storage technologies for applications with time-varying generation and demand, including renewable energy and AI data centers.

The presentation explained an approach that uses power semiconductor switching and timing control to manage energy flows as discrete pulses. A key topic was the dynamic reconfigurable battery network (DRBN), which controls how long individual cells or modules connect to a DC bus. Participants explored how adjusting the participation of batteries with different characteristics can address imbalance and improve usable capacity and operational flexibility.

The diagnostics and operation section introduced online estimation of battery open-circuit voltage and state of charge, together with examples of detecting and isolating faulty modules while continuing operation with the remaining modules. The seminar also discussed large language models (LLMs) and knowledge graphs for fault root cause analysis, battery life prediction, and operation and maintenance support, connecting energy flow control with the use of battery state information.

Industrial examples included a 10 MW/34 MWh energy storage system using retired electric vehicle batteries and experience in deploying and operating large-scale dynamically reconfigurable battery systems. The session also outlined research directions in hierarchical storage architectures combining different storage media and power supply technologies for rapidly changing AI data center loads. These topics helped participants explore the architecture and potential applications of next-generation energy storage systems combining digital control and AI.