Speaker: Keshav Kasturi Rangan, Chemical Engineering and Materials Science, WSU

Time: Wednesday, March 05, from 2:30 pm to 3:30 pm

Location for in-person participants: 1146 FAB 

Zoom link for online audience: https://wayne-edu.zoom.us/j/92845590121?pwd=CpRA5Wa5gzSMn2xiVkR2abD83O5nrH.1

Title: Profit Considerations for Nonlinear Control-Integrated Cyberattack Detection on Process Actuators

Abstract: Smart factories are equipped with control equipment, such as advanced sensors, embedded software, and robotics that collect and analyze data for improved control and decision-making. The integration of these processes with ERP, supply chain, and other enterprise-level systems helps increase visibility of an entire plant and consequently promotes better manufacturing practices. This simultaneously also opens the same industries to cyberattacks affecting all connected processes, such as the attack on the Ukrainian power grid (2015), with debilitating impacts. From a process control literature perspective, strategies to detect cyberattacks on control systems have gained traction in recent years. In accordance with this trend, prior research from our group has focused on the development of several strategies for detecting cyberattacks on process sensors, actuators, or simultaneous attacks on both by modifying an advanced optimization-based control strategy known as Lyapunov-based economic model predictive control (LEMPC). This talk focuses on an active detection strategy for cyberattacks on control components. Active attack detection strategies modify process operation to probe for attacks. Though this may aid with detection, it may also disrupt process operation and potentially reduce profits. The detection strategy was originally formulated to make explicit stability guarantees in the presence of cyberattacks on actuators by perturbing economically optimal process behavior that could for example be implemented using an LEMPC formulation. The active detection strategy continuously probes for cyberattacks using control actions that force the Lyapunov function to decrease by making use of only the contractive constraint of the LEMPC formulation. This formulation changes the state that the controller converges toward at every sampling period and is updated by using a redundant controller that provides optimal states. This formulation allows us to make heuristic strategies geared toward limiting the loss in profits due to active detection. The contribution of this talk is the development of explicit guarantees of profitability in addition to stability by using two LEMPC-based auxiliary controllers to guide a third stabilizing Lyapunov controller. The first of the two redundant controllers, an auxiliary LMPC, uses only the contractive constraint of the LEMPC formulation and is used to benchmark the least profit associated with actual controller, ith LMPC, used to control the process over a sampling period. The second redundant controller, an auxiliary LEMPC (A-LEMPC), is used to determine the states that the ith LMPC should track over every sampling period, consequently enabling them to have the potential to outperform the auxiliary LMPC. This profitability analysis provides a step toward elucidating the conditions under which active detection policies may not result in severe loss in profits. A general process reactor example is used to demonstrate the implementation of the detection strategy.

Bio: Keshav Kasturi Rangan is originally from Bangalore, India. He received his B.E. in Chemical Engineering from R.V. College of Engineering in 2014 and earned an M.S. from Carnegie Mellon University in 2015. After graduation, he worked for three years at startup companies focusing on control solutions (Industrial Learning Systems), automation validation (Panacea Technologies Inc.), and manufacturing operations (Tulip Interfaces). He is currently pursuing a Ph.D. in Chemical Engineering at Wayne State University under the guidance of Dr. Helen Durand. His research interests in the general area of process systems engineering, with a focus on optimal process control strategies and their interaction with quantum computing.

______________________________________________________________________

The CAD Seminar Series is a dedicated platform for advancing knowledge, fostering innovation, and promoting collaboration across the fields of Computation, Artificial Intelligence, and Data Science. This series brings together leading experts, researchers, and professionals to explore the latest developments, tackle emerging challenges, and drive forward-thinking solutions at the convergence of these critical disciplines.

Objectives:

• Advance Knowledge: Share cutting-edge research and insights that push the boundaries of what is known in CAD.

• Foster Innovation: Encourage the development of novel ideas and solutions through interdisciplinary dialogue and creative thinking.

• Promote Collaboration: Unite expertise across disciplines and build bridges between academia, industry, and government to address complex problems and create opportunities for joint ventures.

Target Audience: The CAD Seminar Series is designed for a diverse audience, including faculty, researchers, students, and professionals in Computation, AI, Data Science, and related fields. It serves as a forum for exchanging ideas, networking, and contributing to the growth of these rapidly evolving areas. We highly recommend in-person attendance to enhance engagement and networking opportunities with speakers and fellow participants.

Call for Participation: We welcome contributions from researchers, practitioners, and students. Whether presenting your work, participating in discussions, or attending as a learner, your involvement is crucial to the success of this collaborative initiative.