(721g) Safe Economic Model Predictive Control | AIChE

(721g) Safe Economic Model Predictive Control

Authors 

Ellis, M. - Presenter, University of California, Los Angeles
Christofides, P., University of California, Los Angeles

Maintaining safe operations of process systems is of the highest priority. In fact, it has been argued that dealing with the challenges associated with achieving safe operations should be viewed as a process control problem as process control and automation are ultimately responsible for (safely) regulating a process in the presence of uncertainty [1]. One potential control methodology that may help to integrate safety and control is economic model predictive control (EMPC) [2]-[5] because it is a unique predictive control technique that uses a general cost function in its formulation (i.e., typically, a non-quadratic cost function). Owing to the fact that the objective function does not need to be positive definite with respect to a given set-point, it may incorporate other considerations (outside of feedback control considerations) into the design of the EMPC scheme through the addition of appropriate constraints imposed in the EMPC optimization problem and the modification of the cost function to explicitly account for these other considerations. For instance, an EMPC scheme was formulated for a handling preventive maintenance of control actuators in [5]. Integrating safety and process control is an open and challenging area as new safety metrics, process monitoring methodologies, and control schemes that incorporate safety need to be introduced.

To integrate EMPC with safety considerations (i.e., handle safety through feedback control), a novel control and monitoring methodology is proposed. First, the safeness of a region of operation with respect to a possible failure (i.e., determine if stability of the process can be maintained for any point in the region of operation if a failure occurs) is characterized. Second, a statistical analysis is used to evaluate the probability of a possible failure. To this end, it is important to point out that it might be best with respect to the process economics to operate a process in specific region of operation. However, this region of operation must also be acceptable with respect to the process safety. Even if it is safe to operate in this region, the reliability of process components decreases and the probability of failure increases with time in general [6], and the safeness of this region of operation may decrease with time. Thus, the probability of a potential failure is computed with time. If the probability of a failure exceeds a threshold and the process cannot be safely operated in the region of operation if a failure occurs, the region of operation is shifted to a safer region of operation regardless of the fact that the process economics performance may decrease in this region of operation. To transition between regions of operation, appropriate constraints are added to the EMPC optimization problem. Through a rigorous stability analysis, recursive feasibility and closed-loop stability are analyzed. The proposed control and monitoring technique, which effectively integrates process control, process economics, and safety considerations, is demonstrated with a chemical process example.

[1] Leveson NG, Stephanopoulos G. A system-theoretic, control-inspired view and approach to process safety. AIChE Journal. 2013;60:2-14.
[2] Angeli D, Amrit R, Rawlings JB. On average performance and stability of economic model predictive control. IEEE Transactions on Automatic Control. 2012;57:1615-1626.
[3] Huang R, Harinath E, Biegler LT. Lyapunov stability of economically oriented NMPC for cyclic processes. Journal of Process Control. 2011;21:501-509.
[4] Heidarinejad M, Liu J, Christofides PD. Economic model predictive control of nonlinear process systems using Lyapunov techniques. AIChE Journal. 2012;58:855-870.
[5] Lao L, Ellis M, Christofides PD. Smart Manufacturing: Handling preventive actuator maintenance and economics using model predictive control. AIChE Journal. 2014;60:2179-2196.
[6] Barlow RE, Proschan F. Mathematical Theory of Reliability. Philadelphia, PI: Society for Industrial and Applied Mathematics, 1996.

Topics