The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) Time Series Clustering for Monitoring Fueling Infrastructure Performance with Kalai Ramea - #300
Sep 18, 2019
Kalai Ramea, a data scientist at PARC, specializes in analyzing hydrogen fueling infrastructure. She shares her journey of purchasing a hydrogen car and the crucial research assessing fueling stations. Kalai discusses using temporal clustering to identify usage patterns at these stations, emphasizing the importance of reliability as their numbers grow. Her insights reveal how machine learning can improve station performance and inform policymakers on enhancing the adoption of zero-emission vehicles while addressing the intersection of transportation and energy systems.
AI Snips
Chapters
Transcript
Episode notes
Hydrogen Car Journey
- Kalai Ramea, frustrated by unreliable hydrogen fueling stations, began collecting data on their performance.
- This personal experience led to a research paper analyzing hydrogen station reliability.
Hydrogen Infrastructure
- Home hydrogen fueling is not feasible due to safety concerns and infrastructure limitations.
- Public hydrogen stations are crucial for fuel cell vehicle adoption, similar to electric vehicle charging stations.
Data-Driven Insights
- Combining fuel consumption data with vehicle rebate data reveals trends in station usage.
- This analysis highlighted overstressed and underutilized stations, confirming anecdotal observations.