Process operational safety using model predictive control based on a process Safeness Index

Abstract

It has been repeatedly suggested that the common cause-and-effect approach to evaluating process safety has deficiencies that could be addressed by a systems engineering perspective. A systems approach should consider safety as a system-wide property and thus would be required to integrate all aspects of the process involved with monitoring or manipulating the process dynamics, including the control, alarm, and emergency shut-down systems while operating them independently for redundancy. In this work, we propose initial steps in the first systems safety approach that coordinates the control and safety systems through a common metric (a Safeness Index) and develop a controller formulation that incorporates this index. Specifically, this work presents an economic model predictive control (EMPC) scheme that utilizes a Safeness Index function as a hard constraint to define a safe region of operation termed the safety zone. Under the proposed EMPC design, the closed-loop state of a nonlinear process is guaranteed to enter the safety zone in finite time in the presence of uncertainty while maximizing a stage cost that reflects the economics of the process. Closed-loop stability is established for a nonlinear process under the proposed implementation strategy.