Page 394 - Design of Simple and Robust Process Plants

9.6 Project Methodology for Operation Optimization 381
calculate the maximum exchanger duty in relation to the cooling medium, which
might be subject to the outside temperature. The MBC might use the pressure in
the system as input to a constrained condition caused by heat exchangers. The qual-
ity of product streams might be a constraint, and as such input for the MBC; how-
ever, the value of the constrained quality might also be subject to optimization. The
exercise results in a preliminary balance between MBC and OO activities. During
the project development the borders need to be further defined.

9.6.1.4 Sensitivities ofDOFs on process performance
The sensitivities of the DOFs on process performance set the basics for the econom-
ics of a OO project. The sensitivities are determined quantitatively by perturbations
of the DOFs selected for the optimization of the overall process or process model.
The sensitivities need to be explored over a larger range of the operation, as cur-
rently practiced. Most OO projects are forced against constraints, which are likely to
be outside the existing operational range.

9.6.1.5 Constraints identification
As mentioned earlier, OO will force the process against constraints, and these will
be different under different economical conditions. The constraints are partly deter-
mined during test runs, and partly by calculations, as the processes will not be avail-
able for all extremes of conditions. The most important constraints are capacity con-
straints, particularly at different feed-stocks and product distributions. Be aware that
the maximum operational conditions as experienced by operation personnel are ap-
proximately 5±10% removed from the real constraints. Operations not equipped
with constraint controllers will operate with conform zones that are the challenges
for OO. Capacity constraints are often subject to meteorological conditions; for
example, the daily temperature cycle might be a cheap capacity increment (see Fig-
ure 9.3).

9.6.1.6 Parameters
The parameters that must be estimated or back-calculated during the simulation/
optimization are those which are subject to change during extended operation, and
have a noticeable impact on the economic performance. The obvious parameters
include: compressor efficiencies of larger machines; column efficiencies; reactor
aging; and fouling factors of exchangers as far as they have an impact on process
efficiency or constraints imbedded in the optimization.

9.6.1.7 Control performance
For the implementation of the OO project, a robust controlled operation is a require-
ment. During this step the quality of the control should be monitored and the
amount of upgrading of the basic control layer be determined. An estimate of the
number of MB controllers needs to be prepared based on suboptimizations, con-
straints and decoupling of interactions to be addressed. The analysis will also iden-
tify additional instrumentation to be installed for parameter estimation and perfor-
mance meters.