CURENT Power and Energy Industry Seminar – Mr. Sean Murphy & Mr. Michael Andersen
Date: Friday, September 28, 2018
Time: 12:20 – 1:10pm EST
Location: Room 404, Min H. Kao Building, Knoxville, TN
Title: Architecting a Universal Sensor Analytics Platform for Utilities
Presenter: Mr. Sean Murphy, PingThings, and Mr. Michael Andersen, University of California Berkeley
Abstract: As the number of sensors on the grid continues to grow rapidly, utilities’ inability to deal with the resulting volume, velocity and variety of data limits their efforts to unlock the value in these measurements. A third generation, domain-specific big data platform is required to enable utilities to efficiently and effectively leverage grid data to gain the insights they need to operate successfully in an environment of increasing indeterminacy.
This talk both overviews and discusses the architectural decisions made in creating a platform architected to ingest, store, access, visualize, analyze, and learn from (train machine learning and deep learning algorithms with) data captured by an arbitrary number and type of sensors measuring the grid with nanosecond temporal resolution. Built with open-source tools to prevent vendor lock-in and benefit from the work of thousands of engineers, the platform uses a novel time series database that gives unprecedented ingestion and querying speeds. The platform also handles data quality, generally a major issue in power grid measurements, which enormously improves the ease of working with the data. Furthermore, the platform is designed to enable the rapid prototyping, implementation and deployment of novel analytics, machine learning, and deep learning applications. Finally, the platform’s convenient APIs and open interfaces encourage innovation while its user-facing applications, like the multi-resolution plotter, enable data exploration and the dispersal of results—essential steps in the development of insightful and valuable data use cases.
Bio: Sean Patrick Murphy is the co-CEO of PingThings, Inc., the creators of the universal sensor analytics platform bringing AI to the nation’s electric grid. After earning dual undergraduate degrees with honors in mathematics and electrical engineering from the University of Maryland College Park, Sean completed his graduate work with honors in biomedical engineering at Johns Hopkins University. He stayed on as a senior scientist at the JHU Applied Physics Laboratory for over a decade, where he focused on machine learning, high-performance and cloud-based computing, and anomaly detection. He then completed with Distinction his MBA from Oxford University and built an email analytics startup and a data sciences consulting firm.
Bio: Michael Andersen is an EECS PhD student at the University of California Berkeley working on technologies for smart cities. His research covers wireless sensor networks, time series databases, operating systems for the built environment and secure communication infrastructure. BTrDB originated in his PhD research into scalable analytics on grid data.