(719d) Optimal Actions Prescribed By Model-Predictive Safety

In 2016 [1] we introduced the concept of model-predictive safety (MPS), and in 2017-2020 [2-5] we proposed a computationally-efficient method for the online implementation of MPS, which requires solving min-max optimization problems offline. MPS [1, 4] generates alarm signals that are predictive and systematically account for process nonlinearities and interactions, while typical existing functional safety systems generate reactive, non-interacting alarm signal(s) when a process variable exceeds a threshold. MPS allows for a systematic utilization of dynamic process models to generate predictive alarm signals (alerts) that allow for preventing and mitigating operation hazards proactively in real time.

In this paper, we expand the concept of MPS so that MPS can prescribe optimal, time-invariant, control actions (through overriding controllers) as well as safety actions that prevent imminent and potential (current and future) operation hazards. Min-max optimization problems are formulated and solved to calculate the optimal actions. A nested particle swarm optimization (PSO) parallel algorithm that solves the min-max optimization problems on computers with parallel processors is presented and implemented. The application and performance of the min-max optimization formulations, the PSO algorithm, and MPS, are shown through numerical simulations of a chemical reactor.

References

[1] T.M. Ahooyi, M. Soroush, J.E. Arbogast, W.D. Seider, U.G. Oktem, Model‐predictive safety system for proactive detection of operation hazards, AIChE Journal, 62 (2016) 2024-2042.

[2] Soroush, M., J.E. Arbogast, and W.D. Seider, “Model-Predictive Safety System for Predictive Detection of Operation Hazards: Off-Line Calculation of Most Aggressive Control Actions and Worst-Case Uncertainties,” CAST Division 10 Plenary Session at the 2017 AIChE Annual Meeting, Minneapolis, MN (2017).

[3] Soroush, M., A.A. Shamsabadi, W.D. Seider, and J.E. Arbogast, "Implementation of Model-Predictive Safety Systems to Detect Predictively Operation Hazards in Non-Minimum-Phase Processes," 2018 AIChE Annual Meeting, Pittsburgh, PA (2018).

[4] M. Soroush, L.S. Masooleh, W.D. Seider, U. Oktem, J.E. Arbogast, Model‐predictive safety optimal actions to detect and handle process operation hazards, AIChE Journal, (2020) e16932.

[5] Soroush, M., Samandari Masooleh, L., Oktem, U., Seider, W.D., Arbogast, J.E. “Model-Predictive Safety: Min-Max Optimization to Calculate the Most Aggressive Control Actions and the Worst-Case Uncertainties,” AIChE Annual Meeting, Orlando, FL, November (2019).