Page 368 - Design of Simple and Robust Process Plants

9.4 Performance (Profit) Meter 355

Temperature
24 hrs

Average daily temperature
Time
Capacity

Capacity gain

Average capacity without constraint control
Fig. 9.3. Effect of the day and night cycle on capacity.

least 10±20% of its projected contribution; this means on the order of tenths of a
percent of the operational cost. It is important to realize that:

Any inaccuracy of an optimization activity results in a missed opportunity.

The challenge to achieve this high level of accuracy is often under-estimated. At the
same time, the ultimate contribution of the operational cost reductions are often not
measured. This in contradiction with the capacity performance, which can be
obtained from the capacity performance measurement and often receives much
attention (see also Chapter 10, Section 10.2.1.)
Validation of the optimization activities up to the intended accuracy is not a trivial
task. The factors that play a role in the achievement of this task are:

. Accurate model description, including constraints.
. Parameter estimation (updating on line).
. Steady-state operation for validation of the model, as well as for implementa-
tion in a quasi steady-state process.
. Validation technique based on the minimization of deviations.
. An on-line overall performance measurement.

9.4
Performance (Profit) Meter

In order to quantify the required level of accuracy of optimization, the process must
be provided with an on-line performance measurement (Krist et al., 1994; US Patent
6.038,505). The meter calculates the performance of the process in money terms per