CAREER: Systemic Performance and Robustness Measures for Large-Scale Dynamical Networks

Lehigh University, Bethlehem PA

Investigators

Abstract

Improving energy efficiency as well as robustness to external perturbations in large-scale dynamical networks is critical for long-term sustainability, from engineering infrastructures to living cells. Examples include distributed power networks, distributed emergency response systems, interconnected transportation networks, metabolic pathways, and social and financial networks. The overarching goal of this project is to develop new methodologies to classify viable domain-specific systemic performance, risk, and fragility measures for dynamical networks and quantify inherent fundamental limits on the best achievable values for each class of such systemic measures. The main barrier to achieve this objective is the lack of rigorous knowledge about networks of interconnected dynamical systems when it comes to understanding their structure, dynamics and holistic behaviors. The issue of fundamental limits and their resulting tradeoffs are important for synthesis of dynamical network as they reveal what is achievable, and conversely what is not achievable by feedback control mechanisms. The focus of this project is to reveal important role of underlying dynamical structure and sparse information structure of dynamical networks in emergence of severe fundamental limits on the best achievable levels of performance and robustness. This project is highly relevant for analysis and synthesis of engineered and natural dynamical networks. Recent technological advances in distributed control and dynamical systems cannot be efficiently integrated on real platforms without the necessary theoretical and experimental support. This project will produce data-driven algorithms to assess global performance and robustness properties of engineered dynamical networks by computing the values of viable systemic measures in real-time operation. The application areas include energy, robotic, transportation, and biological networks. This project further seeks to make broad impacts through development and dissemination of courses and material in academia and industry that cover key network science concepts while staying grounded in engineering applications. The technical goal of this project is to develop an integrated theory of dynamical networks based on notions of systemic measures. This project will focus on discovering connections between existing gold standard performance and robustness measures that have been used in disciplines such as control theory, dynamical systems, physics, biology, and finance, and develop a unified systems-theoretic framework for characterization of such measures and their inherent fundamental limits and tradeoffs. This project intends to create a multidisciplinary research environment focusing on developing a foundational science that allows for measuring, predicting, and containing systemic measures for network-oriented applications.