Page 33 - Design of Simple and Robust Process Plants

16 Chapter 1 Introduction
interface, to support the operator in case he or she has to interfere with the opera-
tion.
Robust control design is one of the most difficult tasks, notably because the
requirements for product quality, environmental and safety are increasing, while the
interaction and response times also have a tendency to increase, due to a higher
level of integration. The approach for the design of control configuration is pre-
sented in Chapter 8. In the past, control design was an empirical effort, but current
control strategies and controls must be designed based on dynamic simulations,
and have to meet the criteria of hands-off operation. The design of the control strate-
gies must be based on integral process operation versus unit operation. (Luyben et
al 1998).The design principle followed is that an operator should be able to run the
operation using the basic control layer at specification. Any outage of a higher level
control layer or optimization layer should not result in process outage, although it
might run less optimally. In order to anticipate interactions, it may be necessary to
build hardware and software provisions that uncouple these effects. All of the above
points must be fulfilled if robust control is to be achieved.

1.9
Operation Optimization

Operation optimization has been used for some time now, especially with regard to
scheduling for batch plants, or linear programs for refinery operations.
Currently, off-line programs based on non-linear programming (NLP) techniques
are the state-of-the-art for feedstock evaluation and selection of the most economical
conditions for chemical plants on a daily basis. One precondition however, is that
there is a reasonably accurate reactor model.
Closed loop static optimization for continuous plants based on non-linear tech-
niques is a recent (within the past decade) development that required a robust con-
trol design and an implementation methodology. Forthcoming developments con-
centrate on the dynamic optimization of product type changes and cycle time opti-
mization of batch operations. In this respect, they also include the cycle time
optimization of gradually degrading/fouling systems of continuous plants. The opti-
mization options and methodologies for implementation are discussed in Chap-
ter 9.

1.10
The Efficient Design and Operation of High-quality Process Plants

During the preparation of this book, two questions with a general theme were
received regarding the implementation of simple and robust designs:
1. How can existing processes be improved?