Webinars

UCLA: Intelligent Battery Operation Via AI-Powered Virtual Sensing

Battery management systems (BMS) rely on measurable parameters such as voltage and current to define safe operating limits, yet many catastrophic failure modes – including lithium metal plating – remain undetected. In this webinar, we will introduce the concept of a machine learning–powered virtual reference electrode, which enables real-time prediction of anode voltage without requiring physical reference electrodes or changes in battery hardware. Trained on large datasets from three-electrode cells, this approach allows physically meaningful internal parameters to be inferred using only standard two-electrode measurements. We will show how these virtual sensors enable smart charging protocols that dynamically avoid failure while extending cycle life, and discuss how this framework can generalize to other hard-to-measure battery health indicators.

This webinar will focus on the following key topics:

• Predicting anode voltage without physical sensors
• Enabling smart charging protocols in real time
• Implications for battery safety and lifetime

A PDF copy of the presentation will be sent to all attendees after the event.

Presenter
Dr. Yuzhang Li – Associate Professor at UCLA

Yuzhang Li is an Associate Professor of Chemical and Biomolecular Engineering at UCLA. His research focuses on understanding and controlling battery interfaces through advanced characterization and data-driven approaches. His group develops new experimental and computational tools (including cryogenic electron microscopy and machine learning) to reveal degradation mechanisms and enable safer, higher-performance batteries. Prof. Li’s work spans lithium-ion, lithium-metal, and multivalent battery systems, with an emphasis on translating fundamental insights into practical battery management strategies.

Register Now