Optimizing Product Quality Performance ◾  69
an 18% increase in capacity. One process train involved the
distillation of large quantities of water, and the optimization
reduced energy consumption by $500,000/year. In another
situation, the optimization of distillation operating conditions
resulted in the recovery of additional product from the bottoms that was a waste stream going to an incinerator. The
value of the recovered product was $220,000/year with no
increase in raw material cost or capital. Also, the costs for
transporting and incinerating the waste were reduced.
The largest improvements occurred when the product
from a distillation column was sold out, and another pound
produced meant another pound sold. In one case, the distillation column was being operated at the maximum boilup
rate recommended by the vendor of the distillation trays.
However, the use of small incremental changes by the
sequential optimization routine increased the boilup over a
period of several weeks and resulted in the discovery that
the trays could be operated at 136% of the maximum capacity recommended by the vendor before flooding occurred.
This resulted in additional sales of $800,000 one year during
a peak in demand for the product with no added capital.
In another case, the product from a new world-scale plant
was sold out, and the bottleneck for the plant was the size
of the reboiler on a distillation column. The boilup rate for
the column was being manipulated by a temperature controller. The capacity of the plant was increased by setting
the reboiler to run at its maximum capacity, that is, at the
constraint limit, all of the time, and the control strategy was
changed to use a temperature control loop to manipulate the
reflux flow rate. This resulted in an increase in production
that year of $32 million.
Parkinson 3 reported the use of a system called D-POP
(Distillation Performance Optimization Program) for optimizing the quality performance from distillation columns via
the Internet.