The ultimate goal in the operation of chemical plants is to work at the possible optimal conditions. However, most of the time, a plant is faced with uncertain conditions during its operation. To efficiently handle these uncertainties, chemical plants must have the flexibility to achieve feasible operation over a range of uncertain conditions. One way to accomplish this, is by moving the nominal optimum to some permanently feasible operating point inside the feasible region (back-off point). In a previous study (Bahri et al., 1994), an iterative approach to solve this problem at steady-state has been proposed (Steady State Open-Loop Back- Off Calculation). In order to consider the transient behaviour of a system responding to disturbances, the back-off calculation should be based upon a dynamic model of the system. Having determined the economic penalty associated with the dynamic open-loop back-off, the next step is to estimate the potential recovery that various control schemes of varying complexity might provide. The methodology to solve the Dynamic Back-Off problem and a flowsheet example are presented in this paper.
[1]
J. D. Perkins,et al.
Selection of process control structure based on economics
,
1994
.
[2]
J. E. Cuthrell,et al.
On the optimization of differential-algebraic process systems
,
1987
.
[3]
J. E. Cuthrell,et al.
Simultaneous optimization and solution methods for batch reactor control profiles
,
1989
.
[4]
C. C. Gonzaga,et al.
On Constraint Dropping Schemes and Optimality Functions for a Class of Outer Approximations Algorithms
,
1979
.
[5]
Jeffery S. Logsdon,et al.
Accurate solution of differential-algebraic optimization problems
,
1989
.
[6]
William F. Ames,et al.
Numerical and applied mathematics
,
1989
.