(268d) Next-Generation Manufacturing Considerations: Flow System Cybersecurity, Image-Based Control, Quantum Entanglement, and Learning Proofs

Next-generation manufacturing processes are faced with a range of considerations, ranging from cybersecurity to image-based sensing to automating reasoning. Examples of directions in next-generation manufacturing systems have included designing control systems which attempt to detect cyberattacks [1], utilizing quantum computation as part of scheduling [2], or incorporating novelty detection with convolutional neural networks for image-based control [3]. Motivated by the broad range of topics of relevance to next-generation manufacturing, in this talk, we overview several of our recent results throughout the broad range of next-generation manufacturing-relevant topics.

We first review our recent results related to cybersecurity of control systems for flow in a pipe [4] . Using ANSYS Fluent as a simulator for the flow in the pipe and modifying the flow with a user-defined function, we look at how different sensor placements along the pipe length can impact how quickly an abnormal event (which could be representative of a cyberattack) is noted. Specifically, if a temperature sensor is placed in the middle of the pipe system, we can expect that the time at which the temperature changes can be observed would be halved, as the fluid would take half the time to travel half the pipe. From this view, which involves dealing with process states that vary with respect to time and space, we see that the closer the temperature is measured to the source of the attack, the faster one can signal the abnormality. Though placing more sensors in different locations of the system may have the benefit of fast attack detection, it is important to ask what the minimum security architecture for sensor placement is to detect attacks as fast as possible without significantly adding more costs. In the next part of the talk, we will discuss image-based control, and will review how simulations in Blender can be used to represent image-based controllers [5]. Next, we will discuss why the limitations of quantum entanglement (and specifically that entanglement alone does not permit signaling faster than the speed of light) cannot be overcome through strategies such as state predictions to facilitate useful communication in next-generation manufacturing [6]. Finally, we analyze what the outputs of a language model trained using proofs of Lyapunov-based economic model predictive control (LEMPC) that have been tailed for applications in areas such as cybersecurity of control systems (e.g., [7]). We will discuss what the results are, and why they are the results, to discuss the potential of this method to be used in guiding proofs for process systems engineering-relevant topics.

References:
[1] Narasimhan, S., El‐Farra, N. H., & Ellis, M. J. (2022). A control‐switching approach for cyberattack detection in process systems with minimal false alarms. AIChE Journal, 68(12), e17875.

[2] Ajagekar, A., Al Hamoud, K., & You, F. (2022). Hybrid classical-quantum optimization techniques for solving mixed-integer programming problems in production scheduling. IEEE Transactions on Quantum Engineering, 3, 1-16.

[3] Pulsipher, J. L., Coutinho, L. D., Soderstrom, T. A., & Zavala, V. M. (2022). SAFE-OCC: A novelty detection framework for Convolutional Neural Network sensors and its application in process control. Journal of Process Control, 117, 78-97.

[4] Kasturi Rangan, K., Oyama, H., Azali Assoumani, I., Durand, H., & Ng, K. Y. S. (2023). Cyberphysical Systems and Energy: A Discussion with Reference to an Enhanced Geothermal Process. In Energy Systems and Processes: Recent Advances in Design and Control (pp. 8-1). Melville, New York: AIP Publishing LLC.

[5] Oyama, Henrique, Dominic Messina, Keshav Kasturi Rangan, Akkarakaran Francis Leonard, Kip Nieman, Helen Durand, Katie Tyrrell, Katrina Hinzman, and Michael Williamson. "Development of directed randomization for discussing a minimal security architecture." Digital Chemical Engineering 6 (2023): 100065.

[6] Nieman, K., Kasturi Rangan, K., & Durand, H. (2022). Control Implemented on Quantum Computers: Effects of Noise, Nondeterminism, and Entanglement. Industrial & Engineering Chemistry Research, 61(28), 10133-10155.

[7] Oyama, Henrique, Dominic Messina, Keshav Kasturi Rangan, and Helen Durand. "Lyapunov-based economic model predictive control for detecting and handling actuator and simultaneous sensor/actuator cyberattacks on process control systems." Frontiers in Chemical Engineering 4 (2022): 810129.