Dissertation Talk: An Integrative Approach to Data-Driven Monitoring and Control
Presentation: Dissertation Talk: EE | May 8 | 3-4 p.m. | 540AB Cory Hall
Roel Dobbe, EECS
Ubiquitous computing, algorithms, better devices, connectivity, and the ability to collect, store and probe large amounts of data are all becoming new commodities; commodities that are the key ingredients to enabling new forms of automation and innovation in critical infrastructures.
How can we integrate data-driven techniques to improve and extend the capabilities of existing critical infrastructure, while safeguarding important values such as safety or privacy? In this dissertation, this question forms the driver for modernizing electric distribution networks to deal with higher levels of renewable generation and electrification.
This thesis integrates concepts from control theory, machine learning, optimization, information theory and differential privacy to make concrete contributions in this context:
(1) enabling state estimation for networks with limited sensing, by fusing estimation and load forecasting, (2) decentralizing optimal power flow, by reconstructing the solution to centralized problems with locally available information, (3) providing a formal framework to assess learning-based decentralized policies and determine optimal communication strategies, by adopting a rate distortion approach, and (4) enabling the protection of sensitive information in the objectives and constraints of distributed optimization and control problems, by integrating a differential privacy mechanism.
The dissertation concludes with an overview on the inherent accumulation of bias and error in data-driven decision-making, and a call for assessing one's epistemology and engaging with domain experts and beneficiaries to steer the design of systems towards promoting safe, beneficial and just outcomes.