( about . research . teaching . outreach . personal . calendar )

Big Data Analytics for Societal Scale Cyber-Physical Systems:

Energy Systems

Lillian J. Ratliff, Roy Dong, Henrik Ohlsson, Zico Kolter, Shankar Sastry



In this workshop we bring together researchers and industrial representatives to discuss resilient societal scale cyber physical systems (CPS). We will focus on one of the most rapidly modernizing CPS, the energy sector, and will use tools from a wide range of fields and areas--including automatic control, system identification, compressed sensing, privacy, security, and machine learning--to address potential new technologies for societal-scale CPS.

Description and (Proposed) Schedule



Cyber Physical Systems (CPS) and the Internet of Things are examples of phrases that are being used to describe the integration of a modern information backbone into the functioning of physical systems and infrastructures. We believe that novel algorithms for real time analytics will revolutionize societal scale CPS. The lead adopter of such technologies will be in the energy sector: the smart grid, oil and gas and other industrial infrastructures associated with the energy ecosystem.

The key to providing a substantial advance in analytics is the development of data driven machine learning algorithms that can be scaled to the dimensions required by energy analytics. The large amount of data collected by various sensors has substantial monetary value. For instance, recent studies have reported that the market for energy analytics in the electric grid is estimated to be worth $9.7 billion by 2020. However, highly sophisticated data algorithms are necessary to integrate the data, create data driven models with predictive power, and extract the value and ultimately revenue from the otherwise incomprehensible stream of information.

Consumer choice and the economic model of utilities are being changed radically by analytics. As a result research in developing new analytics also poses challenging, longer term, large scale and interdisciplinary research problems of interest to the areas of automatic control, system identification, compressed sensing, privacy, security and machine learning.

This workshop is intended to be a gathering of individuals from industry and academia (including graduate students, faculty, and researchers) to discuss research in resilient, societal scale cyber-physical systems. The aim is to understand what problems that are of practical relevance and to identify interesting and promising directions for future research.

The workshop will begin with an introductory session in the morning focusing on societal scale cyber-physical systems and analytics. It will be followed by lecture series in four main areas: industrial needs, incentive design, resilience: security, and resilience: privacy. The workshop will conclude with a panel and round-table discussion focusing on the future of research for societal scale cyber-physical systems.

Logistics: 25 minute talks followed by 5 minute discussions

Instructor Bios