(7gx) Discrete and Hybrid Dynamics, Cyber-Physical Systems, and Formal Methods in Chemical Engineering

My research falls into the broad categories of systems, control, cyber-physical systems, and process systems engineering. Specifically, I work on problems related to analyzing and designing the automation logic in a modern chemical plant. This logic is the part of the control system that's responsible for making decisions on the order of every second in response to measurements from the plant and input from operators to ensure that the plant operates safely and efficiently. Some of the key challenges that arise in this field are modeling the behavior of a chemical plant's existing automation logic, specifying the desired behavior of the control and automation system, formally determining whether the existing logic meets the specification, and designing logic that's guaranteed to meet a specification. Each of these challenges is magnified by the large size and high complexity of a typical chemical processing plant.

A cyber-physical system (CPS) is a system that involves interaction between a "cyber" component (such as the control system in a chemical plant) and a "physical" component (such as the chemical process itself). This type of system is not new, but, over roughly the past 10 years, interest in dealing directly with CPSs (as opposed to decomposing the "cyber" and "physical" parts of the systems) has increased significantly in diverse disciplines such as aerospace, computer, electrical, and mechanical engineering. Applications include air-traffic control, software and hardware verification, energy grid management, and self-driving cars. The growth of CPS-related research has not been as strong in chemical and biomedical engineering, despite the fact that examples of CPSs abound on scales from entire chemical plants down to medical devices. The main objective of my research to date, and goal of my future research, is to fill this gap by identifying the tractable problems in the field of chemical engineering that existing CPS theory can be applied to, and by extending the existing theory to address problems that are outside its reach. Such problems that I have worked on (or continue to work on) in my Ph.D. and Postdoctoral research include verification and falsification ("bug finding") of chemical plant automation systems and addressing the interaction between low-level automation logic and high-level scheduling decisions that occur in batch processing plants.

CPS research in chemical engineering will serve to accelerate the application of smart manufacturing and automation (which require a tight integration between physical processes and digital/cyber control and scheduling) in the chemical-processing industry. Smart manufacturing is one of the most promising frontiers in the chemical-processing industry because it seeks to address all or most of the primary manufacturing objectives, including safety, reliability, efficiency, flexibility, and profitability. The research problems that need to be addressed to enable its application include:

• managing discrete/continuous interactions in process control and automation (analysis and design)
• bridging the gap between high-level scheduling decisions (such as how much of a particular product to produce, made by human operators, often with the aid of optimization tools, at time scales on the order of hours and days) and low-level discrete automation decisions (such as whether or not to open a safety valve, made by an embedded control system, at time scales less than a second)
• synthesizing discrete automation logic that is provably correct with respect to a given specification (for example, a safety or operability requirement) when applied in a given plant
• detecting errors in existing automation logic that cause it to violate a given specification

The fact that each of these problems involves CPSs demonstrates the potential impact of pushing forward CPS-related research in chemical engineering.