312338 Optimal Time-Varying Operation of Nonlinear Process Systems With a Two-Layer Economic Model Predictive Control Scheme

Monday, November 4, 2013: 12:30 PM
Continental 8 (Hilton)
Matthew Ellis, Chemical and Biomolecular Engineering, University of California, Los Angeles, Los Angeles, CA, Helen Durand, Chemical and Biomolecular Engineering, University of California, Los Angeles and Panagiotis D. Christofides, Department of Chemical and Biomolecular Engineering, University of California, Los Angeles, Los Angeles, CA

Time-varying operation is not a new concept within chemical process industry and has been extensively studied within the context of periodically operated chemical reactors [1]-[3]. This research has resulted in numerous demonstrations illustrating that periodic operation of various chemical reactors can result in better economic performance. However, optimal time-varying operation has only been previously studied in the context of determination of the optimal periodic switching pattern [3]. Therefore, it is important to develop systematic methods for determining optimal time-varying operating strategies of these various chemical processes. Economic model predictive control can accomplish this goal by determining the economically optimal time-varying operating strategy on-line and in real-time [4]-[6].

In this work, we propose a two-layer approach to dynamic economic optimization and process control for optimal time-varying operation of nonlinear process systems. The upper layer, utilizing a Lyapunov-based economic model predictive control (LEMPC) system, is used to compute dynamic economic optimization policies for process operation. The lower layer, utilizing a Lyapunov-based MPC (LMPC) system, is used to ensure that the closed-loop system state follows the optimal time-varying trajectories computed by the upper layer over each finite-time operating window. To improve the computational efficiency of the two-layer structure, we allow both the LEMPC and the LMPC to compute control actions for two distinct sets of manipulated inputs thus decreasing the real-time computational demand compared to other one-layer EMPC schemes. Following a rigorous formulation and analysis of the proposed method, we demonstrate boundedness of the closed-loop system state and closed-loop economic performance improvement with the proposed two-layer framework compared to steady-state operation as well as with respect to other existing time-varying operation strategies previously proposed in the literature in the context of a benchmark chemical process application. Lastly, we also perform closed-loop simulations of the benchmark chemical process application with other one-layer EMPC structures [4]-[6] and provide a thorough analysis of each control scheme.

References

  1. Silveston P, Hudgins RR, Renken A. Periodic operation of catalytic reactors-introduction and overview. Catalysis Today. 1995;25:91-112.
  2. Budman H, Silveston PL. Control of periodically operated reactors. Chemical Engineering Science. 2008;63:4942-4954.
  3. Ozgulsen F, Adomaitis RA, Cinar A. A numerical method for determining optimal parameter values in forced periodic operation. Chemical Engineering Science. 1992;47:605-613.
  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. Huang R, Harinath E, Biegler LT. Lyapunov stability of economically oriented NMPC for cyclic processes. Journal of Process Control. 2011;21:501-509.
  6. 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.

Extended Abstract: File Not Uploaded
See more of this Session: Networked, Decentralized and Distributed Control
See more of this Group/Topical: Computing and Systems Technology Division