Technical Notes
On machine learning, physical systems, and real-world data

Artificial Intelligence (AI) has been a transformative force across industries, and battery research is no exception. From optimizing materials discovery to real-time monitoring of battery systems, AI promises to accelerate innovation and address some of the most pressing challenges in energy storage. But amidst the buzz, one question persists: Is AI living up to its potential in battery research, or is it more hype than reality? In this blog, we’ll dissect the current state

As a data scientist in the battery industry, I’ve had the privilege to dive deep into the vast ocean of battery data. From controlled lab experiments to real-world applications like electric vehicles (EVs), this data holds the key to unlocking advancements in next-generation battery technology. However, navigating and extracting value from this data presents its own set of challenges and opportunities. The Dual Nature of Battery Data: Challenges and Opportunities Battery data

In our previous blog, we started exploring the key health indicators for assessing the State of Health (SoH) of electric motorbike batteries. We focused on voltage-based features and demonstrated how they can reveal vital information about battery health. In this continuation, we’ll delve into additional health indicators that are crucial for a comprehensive understanding of battery diagnostics. Current Based Features Just like voltage, the current flowing through a battery d