(6id) Optimization-Based Control of Complex Process Networks in Smart Manufacturing: The Appearance of Cyber-Physical Systems, Cloud Computing, and Big Data Analytics

Large-scale process networks are becoming the rule rather than the exception in energy and chemicals production industries due to the increasing demand for integrated and intensified process operation, which provides sustainable productivity improvement benefits. Examples of such systems include chemical processes, power plants, electrical grid, water distribution networks and many more. Process integration and intensification, achieved through energy/material recycling and heat/mass transfer optimization, has been extensively targeted as a means to reduce raw material consumption and lower utility requirements, and thus promote sustainable designs. However, these benefits are accompanied by major challenges in automated operation and control, specifically in smart process manufacturing which requires high levels of control and adaptivity.

In the architecture of the process control systems, different components need to communicate with each other (e.g., the sensors transmit their measurements to the controllers, the controllers use this information to compute the control input, and the control input is then sent to the actuators so that it can be physically implemented). For this communication to take place, a shared communication network is usually deployed across the plant to be controlled. Although wired networks have been traditionally used for this purpose, an increasing number of industrial control systems now use wireless networks since they are easier to deploy and to maintain. In addition, these networks are sometimes connected to the corporate intranet, and in some cases even to the Internet, to utilize off-site and cloud computational resources for system identification (by analyzing a significant amount of process data) and computing control inputs (by solving real-time nonlinear constrained dynamic optimization problems). Consequently, the modern control systems are becoming more open to the cyber-world, and as such, are more vulnerable to attacks that can cause faults and failures. Real-world attacks on control systems have in fact occurred in the past decade and have in some cases caused significant damage to the targeted physical processes, indicate the apparent need for strategies and mechanisms to deal with new security challenges related to the size of the large-scale process networks.

Regarding my future research efforts, I intend to extend my computational work to modeling, identification, and optimization-based estimation and control of large-scale process networks in sustainable and smart manufacturing. Specifically, the intellectual objective of my future research is to develop enabling technologies designed to enhance control and real-time decision making for smart chemical and energy systems in the 4^th generation industrial revolution (Industry 4.0). I aim to bring theoretical tools from cyber-physical systems, internet of things, and cloud computing to solve the communication and real-time computational challenges in networked control of large-scale chemical and energy systems. I plan to use the traditional tools in the big data analytics to address model uncertainty issues. The proposed research will introduce systematic computationally efficient process identification methods to estimate system parameters required by the model predictive control to compute the optimal control action subject to a continuous decrease in the identification error. Note that transferring such large amount of data over the shared communication network must be managed by smart communication protocols and algorithms. Also, my proposed research will cover theoretical and algorithmic aspects of estimation and control of the process networks subject to adversarial attacks and guarantee the maximum resilience of the system through secure distributed control.

Teaching Interests:

I view teaching as an integral part of my overall academic activity. Given my chemical engineering background and my experience as a lecturer, lab supervisor, and teaching assistant in several undergraduate courses, laboratories, and workshops, I feel confident that I can effectively teach every course in the undergraduate chemical engineering curriculum, and I will be happy to do so. I believe that problem-solving is the crucial skill that engineering students should acquire in any class. In my future teaching career, I plan to incorporate active problem-solving in the undergraduate classroom through group discussions and conversation which will engage more students and improve educational outcomes. Given my expertise in the areas of process dynamics, control, design and applied mathematics, my immediate teaching plans concern the improvement and development of computer-aided courses pertaining to process dynamics and control, process design, and analytical/numerical methods for solving chemical engineering problems.

I look forward to strengthening the mathematical and computational content of core chemical engineering courses such as transport phenomena, reaction engineering, and thermodynamics. This strategy will provide the students with required skills for precise modeling of the velocity/pressure fields, heat/mass transfer, equilibrium conditions, and reaction mechanisms of complex systems in real-world applications. It will also supply the necessary tools for the students to develop computationally efficient programs to solve the resulting nonlinear algebraic, ordinary, and partial differential equations. Besides the direct impacts on improving the problem-solving ability in the undergraduate core courses, such an approach can inspire students to employ the power of modern numerical mathematics to provide solutions to cutting-edge research problems across science and engineering. Also, I intend to develop my own graduate-level courses focusing on control and optimization of large-scale process networks, sustainable approach to process control, secure estimation and control of the processes in smart manufacturing, and advanced tools for process data analytics.